In this file, we presented the results from analyses of exposures
against computed epigenetic age accelerations (EAA) calculated using DNA
methylation data by different methods. For each exposure, we conducted
both primary analyses (likelihood ratio tests and generalized estimating
equations (GEE), adjusted for confounders) and sensitivity analyses
(likelihood ratio tests and/or generalized estimating equations (GEE)
limited to certain fuel users, not adjusted for confounders). In
addition to what’s included in the analysis plan, we also analyzed
ambient and urinary exposures.
1.1. Description of study population
There are 129 visits with corresponding epigenetic ages available
among 106 female subjects. For these 106 subjects, 83 have been visited
once and 23 have been visited twice.
The following tables summarize all the information of the first visit
of these 106 subjects.
Baseline characteristics (confounders)
| Characteristic |
N = 106 |
| Age |
56 (15) |
| county |
|
| Fuyuan |
53 / 106 (50%) |
| Xuanwe |
53 / 106 (50%) |
| BMI |
22.0 (3.5) |
| ses |
|
| 0 |
53 / 106 (50%) |
| 1 |
53 / 106 (50%) |
| edu |
|
| 1 |
72 / 106 (68%) |
| 2 |
17 / 106 (16%) |
| 3 |
13 / 106 (12%) |
| 4 |
4 / 106 (3.8%) |
Epigenetic ages
| Characteristic |
N = 106 |
| DNAmAge |
56 (13) |
| DNAmAgeHannum |
59 (14) |
| DNAmPhenoAge |
55 (14) |
| DNAmAgeSkinBloodClock |
56 (13) |
| DNAmGrimAge |
55 (12) |
| DNAmTL |
6.84 (0.33) |
Epigenetic ages accelarations
| Characteristic |
N = 106 |
| AgeAccelerationResidual |
0.2 (4.6) |
| AgeAccelerationResidualHannum |
-0.4 (4.1) |
| AgeAccelPheno |
-0.7 (4.4) |
| DNAmAgeSkinBloodClockAdjAge |
0.0 (3.4) |
| AgeAccelGrim |
-0.32 (2.96) |
| DNAmTLAdjAge |
0.03 (0.18) |
| IEAA |
0.1 (4.3) |
| EEAA |
-0.5 (5.2) |
Fuel/stove type exposures
| Characteristic |
N = 106 |
| curFuel |
|
| Smokeles |
12 / 90 (13%) |
| Smoky |
72 / 90 (80%) |
| Wood_and_or_Plant |
6 / 90 (6.7%) |
| (Missing) |
16 |
| brthFuel |
|
| Mix |
42 / 93 (45%) |
| Smokeles |
3 / 93 (3.2%) |
| Smoky |
40 / 93 (43%) |
| Wood |
8 / 93 (8.6%) |
| (Missing) |
13 |
| cumFuel |
|
| Mix |
64 / 96 (67%) |
| Smoky |
32 / 96 (33%) |
| (Missing) |
10 |
| curStove |
|
| Firepit_and_unventilated |
16 / 90 (18%) |
| Mix |
14 / 90 (16%) |
| Portable_stove |
16 / 90 (18%) |
| Ventilated |
44 / 90 (49%) |
| (Missing) |
16 |
5MC exposures
| Characteristic |
N = 106 |
| cur_5mc |
8.1 (4.1) |
| (Missing) |
2 |
| cum_5mc |
266 (149) |
| (Missing) |
2 |
| bir_5mc |
5.14 (2.81) |
| (Missing) |
2 |
| cur_5mc_measured |
14 (42) |
| (Missing) |
65 |
## [1] "Pearson pair-wise correlation:"
## cur_5mc cum_5mc bir_5mc cur_5mc_measured
## cur_5mc 1.0000000 0.7055284 0.70960439 0.10311689
## cum_5mc 0.7055284 1.0000000 0.84246687 0.17631136
## bir_5mc 0.7096044 0.8424669 1.00000000 -0.04507814
## cur_5mc_measured 0.1031169 0.1763114 -0.04507814 1.00000000
## [1] "Spearman pair-wise correlation:"
## cur_5mc cum_5mc bir_5mc cur_5mc_measured
## cur_5mc 1.0000000 0.6580617 0.6721831 0.4434641
## cum_5mc 0.6580617 1.0000000 0.8313456 0.3335835
## bir_5mc 0.6721831 0.8313456 1.0000000 0.2297288
## cur_5mc_measured 0.4434641 0.3335835 0.2297288 1.0000000
Cluster-based exposures
clusCUR6
Clusters based on model-based exposure estimates at or shortly before
the visit
| Characteristic |
N = 106 |
| CUR6_BC_PAH6 |
0.22 (0.96) |
| (Missing) |
2 |
| CUR6_PAH31 |
0.19 (0.91) |
| (Missing) |
2 |
| CUR6_NkF |
-0.06 (1.09) |
| (Missing) |
2 |
| CUR6_PM_RET |
-0.01 (0.99) |
| (Missing) |
2 |
| CUR6_NO2 |
0.08 (1.01) |
| (Missing) |
2 |
| CUR6_SO2 |
-0.18 (0.88) |
| (Missing) |
2 |
clusCHLD5
Clusters based on model-based exposure estimates accrued before age
18
| Characteristic |
N = 106 |
| CHLD5_X7 |
-0.07 (0.93) |
| (Missing) |
2 |
| CHLD5_X33 |
0.13 (1.00) |
| (Missing) |
2 |
| CHLD5_NkF |
-0.11 (1.10) |
| (Missing) |
2 |
| CHLD5_NO2 |
0.13 (1.11) |
| (Missing) |
2 |
| CHLD5_SO2 |
-0.05 (0.91) |
| (Missing) |
2 |
clusCUM6
Clusters based on model-based lifetime exposure estimates
| Characteristic |
N = 106 |
| CUM6_BC_NO2_PM |
0.03 (1.05) |
| (Missing) |
2 |
| CUM6_PAH36 |
0.14 (0.98) |
| (Missing) |
2 |
| CUM6_DlP |
-0.20 (1.03) |
| (Missing) |
2 |
| CUM6_NkF |
-0.06 (1.09) |
| (Missing) |
2 |
| CUM6_RET |
-0.16 (1.03) |
| (Missing) |
2 |
| CUM6_SO2 |
-0.17 (0.95) |
| (Missing) |
2 |
clusMEAS6
Clusters based on pollutant measurements
| Characteristic |
N = 106 |
| MEAS6_BC_PM_RET |
0.00 (1.01) |
| (Missing) |
57 |
| MEAS6_X31 |
0.02 (1.03) |
| (Missing) |
57 |
| MEAS6_X5 |
0.01 (0.99) |
| (Missing) |
57 |
| MEAS6_DlP |
-0.02 (0.99) |
| (Missing) |
57 |
| MEAS6_NkF |
0.05 (1.04) |
| (Missing) |
57 |
| MEAS6_NO2_SO2 |
0.03 (1.03) |
| (Missing) |
57 |
clusURI5
Clusters based on urinary biomarkers
| Characteristic |
N = 106 |
| URI5_NAP_1M_2M |
0.01 (0.97) |
| (Missing) |
13 |
| URI5_ACE |
-0.12 (0.99) |
| (Missing) |
13 |
| URI5_FLU_PHE |
-0.04 (0.96) |
| (Missing) |
13 |
| URI5_PYR |
-0.06 (0.94) |
| (Missing) |
13 |
| URI5_CHR |
-0.02 (1.03) |
| (Missing) |
13 |
Ambient exposures
| Characteristic |
N = 106 |
| bap_air |
66 (91) |
| (Missing) |
3 |
| pm25_air |
205 (188) |
| ANY_air |
908 (1,545) |
| (Missing) |
33 |
| BPE_air |
69 (93) |
| (Missing) |
3 |
| BaA_air |
91 (153) |
| (Missing) |
3 |
| BbF_air |
110 (151) |
| (Missing) |
3 |
| BkF_air |
24 (33) |
| (Missing) |
3 |
| CHR_air |
88 (141) |
| (Missing) |
3 |
| DBA_air |
23 (36) |
| (Missing) |
3 |
| FLT_air |
65 (146) |
| (Missing) |
3 |
| FLU_air |
441 (691) |
| (Missing) |
33 |
| IPY_air |
41 (50) |
| (Missing) |
3 |
| NAP_air |
5,342 (8,071) |
| (Missing) |
33 |
| PHE_air |
675 (1,079) |
| (Missing) |
33 |
| PYR_air |
71 (149) |
| (Missing) |
3 |
Urinary biomarkers
| Characteristic |
N = 106 |
| Benzanthracene_Chrysene_urine |
0.98 (3.49) |
| (Missing) |
2 |
| Naphthalene_urine |
247 (755) |
| Methylnaphthalene_2_urine |
49 (65) |
| (Missing) |
8 |
| Methylnaphthalene_1_urine |
21 (26) |
| (Missing) |
3 |
| Acenaphthene_urine |
8 (11) |
| Phenanthrene_Anthracene_urine |
216 (296) |
| Fluoranthene_urine |
22 (25) |
| Pyrene_urine |
0.74 (0.62) |
| (Missing) |
15 |
1.2. EAA ~ confounders
## For AgeAccelerationResidual :[1] "Fitting with 129 observations."
## coefficient std ci_lower ci_upper p_val sig_level
## (Intercept) -0.2555 4.8412 -9.7443 9.2334 0.9579 > 0.05
## Age 0.0164 0.0416 -0.0651 0.0979 0.6940 > 0.05
## countyXuanwe -0.3183 0.8670 -2.0175 1.3810 0.7135 > 0.05
## BMI -0.0820 0.1092 -0.2961 0.1321 0.4528 > 0.05
## ses 1.9126 1.2040 -0.4472 4.2724 0.1122 > 0.05
## edu 0.4074 0.5870 -0.7431 1.5580 0.4876 > 0.05
## For AgeAccelerationResidualHannum :[1] "Fitting with 129 observations."
## coefficient std ci_lower ci_upper p_val sig_level
## (Intercept) -4.0613 4.1892 -12.2720 4.1495 0.3323 > 0.05
## Age 0.0447 0.0389 -0.0316 0.1209 0.2508 > 0.05
## countyXuanwe 0.4064 0.7888 -1.1396 1.9523 0.6064 > 0.05
## BMI 0.0047 0.0903 -0.1722 0.1817 0.9584 > 0.05
## ses -0.1213 1.2013 -2.4758 2.2332 0.9196 > 0.05
## edu 0.6429 0.5143 -0.3652 1.6509 0.2113 > 0.05
## For AgeAccelPheno :[1] "Fitting with 129 observations."
## coefficient std ci_lower ci_upper p_val sig_level
## (Intercept) -2.6519 3.9910 -10.4742 5.1704 0.5064 > 0.05
## Age 0.0252 0.0403 -0.0537 0.1042 0.5315 > 0.05
## countyXuanwe -0.4712 0.8431 -2.1237 1.1813 0.5763 > 0.05
## BMI 0.0553 0.0890 -0.1191 0.2298 0.5341 > 0.05
## ses -0.2758 1.0826 -2.3976 1.8461 0.7989 > 0.05
## edu -0.0725 0.5041 -1.0606 0.9155 0.8856 > 0.05
## For DNAmAgeSkinBloodClockAdjAge :[1] "Fitting with 129 observations."
## coefficient std ci_lower ci_upper p_val sig_level
## (Intercept) -1.5906 3.7271 -8.8957 5.7146 0.6696 > 0.05
## Age -0.0053 0.0399 -0.0835 0.0730 0.8954 > 0.05
## countyXuanwe -0.0470 0.6577 -1.3361 1.2421 0.9431 > 0.05
## BMI 0.0846 0.0778 -0.0678 0.2371 0.2766 > 0.05
## ses -0.2236 1.0918 -2.3634 1.9163 0.8377 > 0.05
## edu 0.1747 0.4783 -0.7627 1.1122 0.7149 > 0.05
## For AgeAccelGrim :[1] "Fitting with 129 observations."
## coefficient std ci_lower ci_upper p_val sig_level
## (Intercept) -1.0505 2.4045 -5.7633 3.6623 0.6622 > 0.05
## Age 0.0539 0.0234 0.0080 0.0998 0.0214 <= 0.05
## countyXuanwe -1.0738 0.5590 -2.1694 0.0218 0.0547 > 0.05
## BMI -0.1189 0.0652 -0.2468 0.0089 0.0683 > 0.05
## ses 1.7585 0.7244 0.3387 3.1784 0.0152 <= 0.05
## edu -0.0325 0.2801 -0.5815 0.5164 0.9076 > 0.05
## For DNAmTLAdjAge :[1] "Fitting with 129 observations."
## coefficient std ci_lower ci_upper p_val sig_level
## (Intercept) 0.2668 0.1305 0.0111 0.5226 0.0408 <= 0.05
## Age -0.0030 0.0014 -0.0057 -0.0003 0.0274 <= 0.05
## countyXuanwe 0.0277 0.0351 -0.0411 0.0965 0.4301 > 0.05
## BMI -0.0016 0.0041 -0.0095 0.0063 0.6937 > 0.05
## ses -0.0198 0.0493 -0.1165 0.0769 0.6879 > 0.05
## edu -0.0246 0.0152 -0.0545 0.0052 0.1056 > 0.05
## For IEAA :[1] "Fitting with 129 observations."
## coefficient std ci_lower ci_upper p_val sig_level
## (Intercept) 0.3629 4.8675 -9.1774 9.9031 0.9406 > 0.05
## Age 0.0186 0.0420 -0.0636 0.1009 0.6571 > 0.05
## countyXuanwe -0.6249 0.7946 -2.1824 0.9325 0.4316 > 0.05
## BMI -0.0702 0.1301 -0.3251 0.1848 0.5897 > 0.05
## ses 1.8296 1.2467 -0.6140 4.2732 0.1422 > 0.05
## edu -0.2049 0.5077 -1.2000 0.7902 0.6866 > 0.05
## For EEAA :[1] "Fitting with 129 observations."
## coefficient std ci_lower ci_upper p_val sig_level
## (Intercept) -6.8361 5.1177 -16.8669 3.1947 0.1816 > 0.05
## Age 0.0710 0.0476 -0.0223 0.1643 0.1358 > 0.05
## countyXuanwe 0.6297 1.0042 -1.3387 2.5980 0.5307 > 0.05
## BMI 0.0083 0.1160 -0.2190 0.2357 0.9428 > 0.05
## ses 0.2458 1.4845 -2.6640 3.1555 0.8685 > 0.05
## edu 1.2008 0.6229 -0.0202 2.4218 0.0539 > 0.05
1.2. 5MC ~ confounders
## For cur_5mc :[1] "Fitting with 129 observations."
## coefficient std ci_lower ci_upper p_val sig_level
## (Intercept) 18.2037 4.7051 8.9817 27.4257 0.0001 <= 0.001
## Age -0.0869 0.0348 -0.1551 -0.0187 0.0125 <= 0.05
## countyXuanwe 0.7541 0.7996 -0.8132 2.3213 0.3457 > 0.05
## BMI -0.2026 0.1302 -0.4578 0.0526 0.1196 > 0.05
## ses -1.6191 1.1106 -3.7958 0.5576 0.1449 > 0.05
## edu -0.2050 0.4356 -1.0588 0.6488 0.6379 > 0.05
## For cum_5mc :[1] "Fitting with 129 observations."
## coefficient std ci_lower ci_upper p_val sig_level
## (Intercept) 205.6981 132.0974 -53.2128 464.6090 0.1194 > 0.05
## Age 3.1286 1.0698 1.0318 5.2254 0.0034 <= 0.01
## countyXuanwe 22.2370 26.7781 -30.2481 74.7222 0.4063 > 0.05
## BMI -3.7026 3.9015 -11.3496 3.9444 0.3426 > 0.05
## ses -30.6554 34.5139 -98.3026 36.9919 0.3744 > 0.05
## edu -20.3137 13.3902 -46.5585 5.9311 0.1293 > 0.05
## For bir_5mc :[1] "Fitting with 129 observations."
## coefficient std ci_lower ci_upper p_val sig_level
## (Intercept) 6.3683 1.9366 2.5726 10.1641 0.0010 <= 0.01
## Age -0.0053 0.0217 -0.0478 0.0372 0.8072 > 0.05
## countyXuanwe 0.3936 0.5482 -0.6809 1.4681 0.4728 > 0.05
## BMI -0.0093 0.0294 -0.0669 0.0483 0.7513 > 0.05
## ses -0.0277 0.7455 -1.4889 1.4334 0.9703 > 0.05
## edu -0.6050 0.2810 -1.1558 -0.0542 0.0313 <= 0.05
## For cur_5mc_measured :[1] "Fitting with 129 observations."
## coefficient std ci_lower ci_upper p_val sig_level
## (Intercept) 22.7197 33.0223 -42.0041 87.4434 0.4914 > 0.05
## Age 0.1417 0.2671 -0.3818 0.6651 0.5958 > 0.05
## countyXuanwe 19.2364 15.8034 -11.7384 50.2111 0.2235 > 0.05
## BMI -1.2999 1.3304 -3.9075 1.3078 0.3286 > 0.05
## ses -10.3530 12.6056 -35.0601 14.3540 0.4115 > 0.05
## edu 4.7439 5.9214 -6.8620 16.3499 0.4230 > 0.05
1.3. clusCUR6 ~ confounders
## For CUR6_BC_PAH6 :[1] "Fitting with 129 observations."
## coefficient std ci_lower ci_upper p_val sig_level
## (Intercept) 0.1220 0.8080 -1.4617 1.7056 0.8800 > 0.05
## Age 0.0056 0.0073 -0.0087 0.0200 0.4413 > 0.05
## countyXuanwe 0.6166 0.1783 0.2670 0.9662 0.0005 <= 0.001
## BMI -0.0183 0.0242 -0.0658 0.0292 0.4496 > 0.05
## ses 0.1067 0.2137 -0.3122 0.5256 0.6176 > 0.05
## edu -0.1204 0.1175 -0.3507 0.1098 0.3053 > 0.05
## For CUR6_PAH31 :[1] "Fitting with 129 observations."
## coefficient std ci_lower ci_upper p_val sig_level
## (Intercept) 2.0273 0.9878 0.0912 3.9635 0.0401 <= 0.05
## Age -0.0135 0.0080 -0.0291 0.0022 0.0920 > 0.05
## countyXuanwe 0.2204 0.1783 -0.1291 0.5699 0.2165 > 0.05
## BMI -0.0445 0.0280 -0.0994 0.0104 0.1121 > 0.05
## ses -0.2465 0.2252 -0.6880 0.1950 0.2738 > 0.05
## edu -0.0588 0.1113 -0.2769 0.1593 0.5973 > 0.05
## For CUR6_NkF :[1] "Fitting with 129 observations."
## coefficient std ci_lower ci_upper p_val sig_level
## (Intercept) 1.9227 1.0706 -0.1757 4.0211 0.0725 > 0.05
## Age -0.0159 0.0091 -0.0338 0.0020 0.0817 > 0.05
## countyXuanwe -0.1341 0.2108 -0.5473 0.2792 0.5249 > 0.05
## BMI -0.0460 0.0337 -0.1120 0.0200 0.1720 > 0.05
## ses -0.1538 0.2925 -0.7272 0.4195 0.5990 > 0.05
## edu 0.0429 0.1330 -0.2178 0.3036 0.7469 > 0.05
## For CUR6_PM_RET :[1] "Fitting with 129 observations."
## coefficient std ci_lower ci_upper p_val sig_level
## (Intercept) 1.0848 0.8239 -0.5300 2.6996 0.1879 > 0.05
## Age -0.0026 0.0105 -0.0233 0.0180 0.8016 > 0.05
## countyXuanwe -0.3492 0.1920 -0.7255 0.0271 0.0689 > 0.05
## BMI -0.0296 0.0203 -0.0695 0.0102 0.1450 > 0.05
## ses -0.0696 0.2933 -0.6444 0.5053 0.8125 > 0.05
## edu -0.0561 0.0955 -0.2434 0.1311 0.5567 > 0.05
## For CUR6_NO2 :[1] "Fitting with 129 observations."
## coefficient std ci_lower ci_upper p_val sig_level
## (Intercept) 0.5912 0.8152 -1.0066 2.1890 0.4683 > 0.05
## Age 0.0010 0.0077 -0.0142 0.0161 0.8997 > 0.05
## countyXuanwe -0.7820 0.1846 -1.1438 -0.4202 0.0000 <= 0.001
## BMI 0.0004 0.0231 -0.0449 0.0457 0.9866 > 0.05
## ses -0.2588 0.2266 -0.7030 0.1855 0.2536 > 0.05
## edu -0.0302 0.1412 -0.3069 0.2466 0.8307 > 0.05
## For CUR6_SO2 :[1] "Fitting with 129 observations."
## coefficient std ci_lower ci_upper p_val sig_level
## (Intercept) 0.4673 0.7815 -1.0644 1.9990 0.5499 > 0.05
## Age -0.0071 0.0091 -0.0249 0.0106 0.4308 > 0.05
## countyXuanwe -0.4356 0.1658 -0.7605 -0.1108 0.0086 <= 0.01
## BMI 0.0033 0.0097 -0.0157 0.0222 0.7365 > 0.05
## ses -0.3479 0.2257 -0.7903 0.0945 0.1232 > 0.05
## edu 0.0512 0.1278 -0.1994 0.3017 0.6890 > 0.05
2.1. Current (self-reported) fuel type
The numbers of observations with each current fuel type:
##
## Smokeles Smoky Wood_and_or_Plant
## 17 87 8
Primary analysis
Investigate the association with current (self-reported) fuel type in
the LEX study participants, adjusting for known confounders. The
reference group for this analysis would be the smoky coal users. This
would be a categorical analysis, and the results would be a p-value from
the likelihood ratio (LR) test of a confounder-only model to a model
including the exposure variables, as well as p-values for the contrast
of each category of coal use (smokeless coal or plant/wood) to that of
smoky coal. FDR correction should be used separately for each of these
sets. The main interest would be in the coal-specific findings and
perhaps less so in the results from the LR test.
Likelihood ratio (LR) test (mix model)
Full model: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * I(\text{Smokeles}) + \beta_2 *
I(\text{Wood_and_or_Plant}) \\
& + \beta_3 * county + \beta_4 * BMI + \beta_5 * ses + \beta_6 *
edu + \epsilon
\end{aligned}
\] Nested model: \[
\begin{aligned}
Y = & \beta_0 \\
& + \beta_1 * county + \beta_2 * BMI + \beta_3 * ses + \beta_4 *
edu + \epsilon
\end{aligned}
\] \(H_0\): The full model and
the nested model fit the data equally well. Thus, you should use the
nested model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.2094 0.6181
## Hannum EAA 0.5417 0.6181
## PhenoAge EAA 0.5333 0.6181
## Skin&Blood EAA 0.4234 0.6181
## GrimAge EAA 0.0369 0.2952
## DNAmTL 0.4400 0.6181
## IEAA 0.4339 0.6181
## EEAA 0.6181 0.6181
GEE (Mix)
In this section, we perform the generalized estimating equations
(GEE) to evaluate the associations between current fuel type and each
Epigenetic Age Acceleration with the formula: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * I(\text{Smokeles}) + \beta_2 *
I(\text{Wood_and_or_Plant}) \\
& + \beta_3 * county + \beta_4 * BMI + \beta_5 * ses + \beta_6 *
edu + \epsilon
\end{aligned}
\]
Results:
## [1] "Fitting with 112 observations."
## [1] "Fitting with 112 observations."
## [1] "Fitting with 112 observations."
## [1] "Fitting with 112 observations."
## [1] "Fitting with 112 observations."
## [1] "Fitting with 112 observations."
## [1] "Fitting with 112 observations."
## [1] "Fitting with 112 observations."
## [1] "Fitting with 112 observations."
## The estimated average GrimAge EAA difference of current fuel type, smokeless/wood or plant compared to smoky coal:
## coefficient std ci_lower ci_upper p_val
## Smoky (reference/intercept) 4.3011 1.9267 0.5247 8.0775 0.0256
## Smokeles -1.7662 0.7112 -3.1601 -0.3723 0.0130
## Wood_and_or_Plant 0.3677 1.4074 -2.3909 3.1263 0.7939
## sig_level
## Smoky (reference/intercept) <= 0.05
## Smokeles <= 0.05
## Wood_and_or_Plant > 0.05
Sensitivity analyses
Likelihood ratio (LR) test (no confounders)
Limit the analyses in the primary analysis to include only a single
observation from each subject (no need for a mixed model). The rationale
for this is that it is not so easy to obtain unbiased p-values from a
mixed model for FDR testing. This can be remediated during FDR testing
but would be good to check.
Full model: \[Y = \beta_0 + \beta_1 *
I(\text{Smokeles}) + \beta_2 * I(\text{Wood_and_or_Plant}) +
\epsilon\] Nested model: \[Y = \beta_0
+ \epsilon\] \(H_0\): The full
model and the nested model fit the data equally well. Thus, you should
use the nested model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.3963 0.8499
## Hannum EAA 0.5312 0.8499
## PhenoAge EAA 0.9482 0.9482
## Skin&Blood EAA 0.5138 0.8499
## GrimAge EAA 0.1984 0.8499
## DNAmTL 0.8784 0.9482
## IEAA 0.4591 0.8499
## EEAA 0.7273 0.9482
GEE (no confounders)
In this section, we perform the generalized estimating equations
(GEE) to evaluate the associations between current fuel type and each
Epigenetic Age Acceleration with the formula: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * I(\text{Smokeles}) + \beta_2 *
I(\text{Wood_and_or_Plant}) + \epsilon
\end{aligned}
\]
Results:
## [1] "Fitting with 112 observations."
## [1] "Fitting with 112 observations."
## [1] "Fitting with 112 observations."
## [1] "Fitting with 112 observations."
## [1] "Fitting with 112 observations."
## [1] "Fitting with 112 observations."
## [1] "Fitting with 112 observations."
## [1] "Fitting with 112 observations."
## [1] "Fitting with 112 observations."
## The estimated average GrimAge EAA difference of current fuel type, smokeless/wood or plant compared to smoky coal:
## coefficient std ci_lower ci_upper p_val
## Smoky (reference/intercept) -0.3573 0.3122 -0.9692 0.2546 0.2524
## Smokeles -1.1566 0.5696 -2.2731 -0.0401 0.0423
## Wood_and_or_Plant 0.1442 1.2705 -2.3460 2.6344 0.9096
## sig_level
## Smoky (reference/intercept) > 0.05
## Smokeles <= 0.05
## Wood_and_or_Plant > 0.05
2.2. Cumulative lifetime (self-reported) fuel type
The numbers of observations with each cumulative lifetime fuel
type:
##
## Mix Smoky
## 82 37
Primary analysis
Likelihood ratio (LR) test (mix model)
Full model: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * I(\text{Mix}) \\
& + \beta_2 * county + \beta_3 * BMI + \beta_4 * ses + \beta_5 *
edu + \epsilon
\end{aligned}
\] Nested model: \[
\begin{aligned}
Y = & \beta_0 \\
& + \beta_1 * county + \beta_2 * BMI + \beta_3 * ses + \beta_4 *
edu + \epsilon
\end{aligned}
\] \(H_0\): The full model and
the nested model fit the data equally well. Thus, you should use the
nested model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.1767 0.4712
## Hannum EAA 0.7839 0.8959
## PhenoAge EAA 0.6268 0.8959
## Skin&Blood EAA 0.9039 0.9039
## GrimAge EAA 0.0549 0.2236
## DNAmTL 0.4661 0.8959
## IEAA 0.0559 0.2236
## EEAA 0.7412 0.8959
GEE (Mix)
In this section, we perform the generalized estimating equations
(GEE) to evaluate the associations between cumulative fuel type and each
Epigenetic Age Acceleration with the formula: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * I(\text{Mix}) \\
& + \beta_2 * county + \beta_3 * BMI + \beta_4 * ses + \beta_5 *
edu + \epsilon
\end{aligned}
\]
Results:
## [1] "Fitting with 119 observations."
## [1] "Fitting with 119 observations."
## [1] "Fitting with 119 observations."
## [1] "Fitting with 119 observations."
## [1] "Fitting with 119 observations."
## [1] "Fitting with 119 observations."
## [1] "Fitting with 119 observations."
## [1] "Fitting with 119 observations."
## [1] "Fitting with 119 observations."
## The estimated average GrimAge EAA difference of cumulative fuel type, mix compared to smoky coal:
## coefficient std ci_lower ci_upper p_val
## Smoky (reference/intercept) 3.9551 1.7563 0.5128 7.3975 0.0243
## Mix -1.0455 0.5286 -2.0816 -0.0095 0.0479
## sig_level
## Smoky (reference/intercept) <= 0.05
## Mix <= 0.05
Sensitivity analyses
Likelihood ratio (LR) test (no confounders)
Full model: \[Y = \beta_0 + \beta_1 *
I(\text{Mix}) + \epsilon\]
Nested model: \[Y = \beta_0 +
\epsilon\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.3878 0.8979
## Hannum EAA 0.7073 0.8979
## PhenoAge EAA 0.8979 0.8979
## Skin&Blood EAA 0.8313 0.8979
## GrimAge EAA 0.1501 0.6004
## DNAmTL 0.6288 0.8979
## IEAA 0.0851 0.6004
## EEAA 0.5624 0.8979
GEE (no confounders)
In this section, we perform the generalized estimating equations
(GEE) to evaluate the associations between cumulative fuel type and each
Epigenetic Age Acceleration with the formula: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * I(\text{Mix}) + \epsilon
\end{aligned}
\]
Results:
## [1] "Fitting with 119 observations."
## [1] "Fitting with 119 observations."
## [1] "Fitting with 119 observations."
## [1] "Fitting with 119 observations."
## [1] "Fitting with 119 observations."
## [1] "Fitting with 119 observations."
## [1] "Fitting with 119 observations."
## [1] "Fitting with 119 observations."
## [1] "Fitting with 119 observations."
## The estimated average GrimAge EAA difference of cumulative fuel type, mix compared to smoky coal:
## coefficient std ci_lower ci_upper p_val
## Smoky (reference/intercept) 0.0399 0.4032 -0.7504 0.8303 0.9211
## Mix -0.7626 0.5444 -1.8296 0.3045 0.1613
## sig_level
## Smoky (reference/intercept) > 0.05
## Mix > 0.05
2.3. Childhood (self-reported) fuel type
The numbers of observations with each current fuel type:
##
## Mix Smokeles Smoky Wood
## 53 5 47 11
Primary analysis
Likelihood ratio (LR) test (mix model)
Full model: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * I(\text{Wood}) + \beta_2 *
I(\text{Smokeles}) + \beta_3 * I(\text{Mix}) \\
& + \beta_4 * county + \beta_5 * BMI + \beta_6 * ses + \beta_7 *
edu + \epsilon
\end{aligned}
\] Nested model: \[
\begin{aligned}
Y = & \beta_0 \\
& + \beta_1 * county + \beta_2 * BMI + \beta_3 * ses + \beta_4 *
edu + \epsilon
\end{aligned}
\] \(H_0\): The full model and
the nested model fit the data equally well. Thus, you should use the
nested model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.1712 0.4565
## Hannum EAA 0.3407 0.5451
## PhenoAge EAA 0.7569 0.7569
## Skin&Blood EAA 0.7251 0.7569
## GrimAge EAA 0.0076 0.0608
## DNAmTL 0.5379 0.7172
## IEAA 0.0742 0.2968
## EEAA 0.3041 0.5451
GEE (Mix)
In this section, we perform the generalized estimating equations
(GEE) to evaluate the association between current fuel type and the Grim
Epigenetic Age Acceleration with the formula: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * I(\text{Wood}) + \beta_2 *
I(\text{Smokeles}) + \beta_3 * I(\text{Mix}) \\
& + \beta_4 * county + \beta_5 * BMI + \beta_6 * ses + \beta_7 *
edu + \epsilon
\end{aligned}
\]
Results:
## [1] "Fitting with 116 observations."
## [1] "Fitting with 116 observations."
## [1] "Fitting with 116 observations."
## [1] "Fitting with 116 observations."
## [1] "Fitting with 116 observations."
## [1] "Fitting with 116 observations."
## [1] "Fitting with 116 observations."
## [1] "Fitting with 116 observations."
## [1] "Fitting with 116 observations."
## The estimated average GrimAge EAA difference of childhood fuel type, smokeless/wood/mixed compared to smoky coal:
## coefficient std ci_lower ci_upper p_val
## Smoky (reference/intercept) 3.4908 1.7706 0.0205 6.9611 0.0487
## Wood 0.1337 0.8601 -1.5521 1.8196 0.8764
## Smokeles -3.9193 1.2197 -6.3098 -1.5287 0.0013
## Mix -1.5379 0.5532 -2.6221 -0.4537 0.0054
## sig_level
## Smoky (reference/intercept) <= 0.05
## Wood > 0.05
## Smokeles <= 0.01
## Mix <= 0.01
Sensitivity analyses
Likelihood ratio (LR) test (no confounders)
Limit the analyses in the primary analysis to include only a single
observation from each subject (no need for a mixed model). The rationale
for this is that it is not so easy to obtain unbiased p-values from a
mixed model for FDR testing. This can be remediated during FDR testing
but would be good to check.
Full model: \[Y = \beta_0 + \beta_1 *
I(\text{Wood}) + \beta_2 * I(\text{Smokeles}) + \beta_3 * I(\text{Mix})
+ \epsilon\] Nested model: \[Y =
\beta_0 + \epsilon\] \(H_0\):
The full model and the nested model fit the data equally well. Thus, you
should use the nested model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.2625 0.5952
## Hannum EAA 0.3720 0.5952
## PhenoAge EAA 0.8202 0.8202
## Skin&Blood EAA 0.7436 0.8202
## GrimAge EAA 0.0098 0.0784
## DNAmTL 0.6013 0.8017
## IEAA 0.1126 0.4504
## EEAA 0.3195 0.5952
GEE (No confounders)
In this section, we perform the generalized estimating equations
(GEE) to evaluate the association between current fuel type and the Grim
Epigenetic Age Acceleration with the formula: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * I(\text{Wood}) + \beta_2 *
I(\text{Smokeles}) + \beta_3 * I(\text{Mix}) + \epsilon
\end{aligned}
\]
Results:
## [1] "Fitting with 116 observations."
## [1] "Fitting with 116 observations."
## [1] "Fitting with 116 observations."
## [1] "Fitting with 116 observations."
## [1] "Fitting with 116 observations."
## [1] "Fitting with 116 observations."
## [1] "Fitting with 116 observations."
## [1] "Fitting with 116 observations."
## [1] "Fitting with 116 observations."
## The estimated average GrimAge EAA difference of childhood fuel type, smokeless/wood/mixed compared to smoky coal:
## coefficient std ci_lower ci_upper p_val
## Smoky (reference/intercept) 0.3204 0.3936 -0.4511 1.0919 0.4157
## Wood 0.0987 0.9388 -1.7414 1.9389 0.9162
## Smokeles -3.8787 1.3488 -6.5224 -1.2350 0.0040
## Mix -1.4468 0.5851 -2.5936 -0.3001 0.0134
## sig_level
## Smoky (reference/intercept) > 0.05
## Wood > 0.05
## Smokeles <= 0.01
## Mix <= 0.05
2.4. Current and cumulative fuel type
Fill the missing current and cumulative fuel type as
Other.
The numbers of observations with each fuel type:
##
## Other Smoky Mix
## Other 10 2 5
## Smoky 0 35 52
## Smokeles 0 0 17
## Wood_and_or_Plant 0 0 8
Primary analysis
GEE (Mix)
In this section, we perform the generalized estimating equations
(GEE) to evaluate the associations between current fuel type and each
Epigenetic Age Acceleration with the formula: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * I(\text{cur_Smokeles}) + \beta_2 *
I(\text{cur_Smoky}) + \beta_3 * I(\text{cur_Wood_and_or_Plant}) \\
& + \beta_4 * I(\text{cum_Smoky})+ \beta_5* I(\text{cum_Mix}) \\
& + \beta_6 * county + \beta_7 * BMI + \beta_8 * ses + \beta_{9} *
edu + \epsilon
\end{aligned}
\]
Results:
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
Sensitivity analyses
GEE (No confounders)
In this section, we perform the generalized estimating equations
(GEE) to evaluate the associations between current fuel type and each
Epigenetic Age Acceleration with the formula: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * I(\text{cur_Smokeles}) + \beta_2 *
I(\text{cur_Smoky}) + \beta_3 * I(\text{cur_Wood_and_or_Plant}) \\
& + \beta_4 * I(\text{cum_Smoky})+ \beta_5* I(\text{cum_Mix}) \\
& + \epsilon
\end{aligned}
\]
Results:
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
3.1. Clusters based on model-based exposure estimates at or shortly
before the visit (clusCUR6)
The file “LEX_clusCUR6.csv” has information on current pollutant
exposures, obtained for the 2 years preceding the visit. To reduce
multi-collinearity between exposures, exposure prototypes were derived
based on hierarchical cluster analysis in combination followed by
principal components analysis. These estimates are available for 6
different prototypes (cluster variables) for a total of 161 subjects and
211 visits. The prototypes are labelled as:
CUR6_BC_PAH6 – Black carbon (BC) and 6 PAHs
CUR6_PAH31 – a large cluster of 31 PAHs
CUR6_NkF – NkF only
CUR6_PM_RET – Particulate matter (PM) and retene
CUR6_NO2 – NO2 only
CUR6_SO2 – SO2 only
Summary the exposure estimates:
| Characteristic |
N = 129 |
| CUR6_BC_PAH6 |
0.79 (-0.5, 0.8) |
| (Missing) |
3 |
| CUR6_PAH31 |
0.41 (-0.4, 0.6) |
| (Missing) |
3 |
| CUR6_NkF |
-0.33 (-0.6, 0.7) |
| (Missing) |
3 |
| CUR6_PM_RET |
-0.31 (-0.5, 0.4) |
| (Missing) |
3 |
| CUR6_NO2 |
0.02 (-0.5, 0.8) |
| (Missing) |
3 |
| CUR6_SO2 |
-0.30 (-0.9, 0.2) |
| (Missing) |
3 |
## By current fuel type:
| Characteristic |
Overall, N = 112 |
Smokeles, N = 17 |
Smoky, N = 87 |
Wood_and_or_Plant, N = 8 |
| CUR6_BC_PAH6 |
0.79 (-0.5, 0.8) |
-1.32 (-1.4, -0.9) |
0.80 (-0.2, 1.1) |
0.69 (0.1, 0.7) |
| (Missing) |
3 |
2 |
1 |
0 |
| CUR6_PAH31 |
0.38 (-0.4, 0.6) |
-1.14 (-1.4, -0.5) |
0.46 (-0.1, 0.6) |
0.75 (0.4, 0.8) |
| (Missing) |
3 |
2 |
1 |
0 |
| CUR6_NkF |
-0.40 (-0.6, 0.7) |
0.06 (-0.2, 0.3) |
-0.51 (-0.6, 0.9) |
0.74 (-0.2, 0.7) |
| (Missing) |
3 |
2 |
1 |
0 |
| CUR6_PM_RET |
-0.32 (-0.5, 0.4) |
-0.04 (-0.9, 0.3) |
-0.32 (-0.5, 0.1) |
2.49 (0.9, 2.6) |
| (Missing) |
3 |
2 |
1 |
0 |
| CUR6_NO2 |
0.06 (-0.4, 0.8) |
1.00 (0.6, 1.4) |
-0.06 (-0.5, 0.5) |
0.63 (-0.2, 1.3) |
| (Missing) |
3 |
2 |
1 |
0 |
| CUR6_SO2 |
-0.30 (-0.9, 0.3) |
1.37 (0.2, 1.5) |
-0.30 (-0.9, 0.1) |
-1.00 (-1.3, -0.9) |
| (Missing) |
3 |
2 |
1 |
0 |
Primary analysis
Likelihood ratio (LR) test (mix model)
Full model: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * \text{BC_PAH6} + \beta_2 * \text{PAH31}
+ \beta_3 * \text{NkF} + \beta_4 * \text{PM_RET} + \beta_5 * \text{NO2}
+ \beta_6 * \text{SO2}\\
& + \beta_7 * county + \beta_8 * BMI + \beta_9 * ses + \beta_{10}
* edu + \epsilon
\end{aligned}
\] Nested model: \[
\begin{aligned}
Y = & \beta_0 \\
& + \beta_1 * county + \beta_2 * BMI + \beta_3 * ses + \beta_4 *
edu + \epsilon
\end{aligned}
\] \(H_0\): The full model and
the nested model fit the data equally well. Thus, you should use the
nested model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.1748 0.3496
## Hannum EAA 0.4741 0.4741
## PhenoAge EAA 0.0403 0.1117
## Skin&Blood EAA 0.0419 0.1117
## GrimAge EAA 0.0276 0.1117
## DNAmTL 0.3324 0.4432
## IEAA 0.4494 0.4741
## EEAA 0.2849 0.4432
GEE (Mix)
In this section, we perform the generalized estimating equations
(GEE) to evaluate the associations between each cluster within current
pollutant exposures and each Epigenetic Age Acceleration with the
formula: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * X \\
& + \beta_2 * county + \beta_3 * BMI + \beta_4 * ses + \beta_5 *
edu + \epsilon
\end{aligned}
\] where \(Y\) is one of the
epigenetic age accelerations and X is one of the cluster estimates.
Results:
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## The estimated effects:
## $CUR6_BC_PAH6
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.4376 0.5530 -1.5214 0.6463 0.4288
## AgeAccelerationResidualHannum -0.3263 0.4319 -1.1727 0.5201 0.4499
## AgeAccelPheno 0.1699 0.4560 -0.7239 1.0637 0.7094
## DNAmAgeSkinBloodClockAdjAge -0.1054 0.3570 -0.8051 0.5942 0.7677
## AgeAccelGrim 0.7280 0.2647 0.2091 1.2468 0.0060
## DNAmTLAdjAge 0.0197 0.0208 -0.0210 0.0604 0.3438
## IEAA -0.1731 0.4729 -1.0999 0.7538 0.7144
## EEAA -0.6292 0.5447 -1.6968 0.4383 0.2480
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim <= 0.01
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUR6_PAH31
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.3157 0.5216 -0.7067 1.3382 0.5450
## AgeAccelerationResidualHannum -0.1466 0.4576 -1.0436 0.7504 0.7487
## AgeAccelPheno 0.4073 0.4871 -0.5475 1.3620 0.4031
## DNAmAgeSkinBloodClockAdjAge 0.3129 0.3496 -0.3722 0.9981 0.3707
## AgeAccelGrim 0.8305 0.2287 0.3821 1.2788 0.0003
## DNAmTLAdjAge -0.0104 0.0157 -0.0411 0.0204 0.5090
## IEAA 0.2367 0.5351 -0.8121 1.2855 0.6582
## EEAA -0.2610 0.5683 -1.3748 0.8528 0.6461
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim <= 0.001
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUR6_NkF
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.1266 0.4508 -1.0101 0.7570 0.7789
## AgeAccelerationResidualHannum -0.1760 0.4085 -0.9766 0.6246 0.6665
## AgeAccelPheno -0.4256 0.3837 -1.1778 0.3265 0.2673
## DNAmAgeSkinBloodClockAdjAge -0.0851 0.3348 -0.7414 0.5711 0.7993
## AgeAccelGrim 0.2463 0.2348 -0.2139 0.7066 0.2942
## DNAmTLAdjAge -0.0286 0.0167 -0.0613 0.0041 0.0870
## IEAA -0.1414 0.3501 -0.8276 0.5447 0.6862
## EEAA -0.1996 0.5345 -1.2472 0.8480 0.7088
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUR6_PM_RET
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.2394 0.4203 -0.5843 1.0632 0.5689
## AgeAccelerationResidualHannum -0.4501 0.3633 -1.1622 0.2620 0.2154
## AgeAccelPheno -0.4008 0.5463 -1.4716 0.6700 0.4632
## DNAmAgeSkinBloodClockAdjAge -0.3933 0.4285 -1.2331 0.4465 0.3586
## AgeAccelGrim 0.4174 0.3269 -0.2234 1.0582 0.2017
## DNAmTLAdjAge 0.0051 0.0206 -0.0353 0.0456 0.8032
## IEAA 0.3620 0.3769 -0.3767 1.1008 0.3368
## EEAA -0.5788 0.5099 -1.5782 0.4206 0.2563
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUR6_NO2
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.6277 0.4601 -0.2741 1.5295 0.1725
## AgeAccelerationResidualHannum -0.1788 0.3989 -0.9607 0.6030 0.6539
## AgeAccelPheno 0.0840 0.5150 -0.9255 1.0935 0.8704
## DNAmAgeSkinBloodClockAdjAge 0.2779 0.3567 -0.4213 0.9771 0.4360
## AgeAccelGrim -0.2031 0.2992 -0.7895 0.3832 0.4971
## DNAmTLAdjAge 0.0241 0.0186 -0.0123 0.0605 0.1937
## IEAA 0.4382 0.4023 -0.3503 1.2268 0.2761
## EEAA -0.2345 0.5714 -1.3543 0.8854 0.6815
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUR6_SO2
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.2692 0.4852 -1.2201 0.6818 0.5790
## AgeAccelerationResidualHannum -0.3495 0.5162 -1.3612 0.6622 0.4983
## AgeAccelPheno -0.7417 0.5946 -1.9072 0.4237 0.2122
## DNAmAgeSkinBloodClockAdjAge -0.4359 0.4445 -1.3070 0.4353 0.3268
## AgeAccelGrim -0.6161 0.3067 -1.2173 -0.0149 0.0446
## DNAmTLAdjAge 0.0017 0.0201 -0.0376 0.0411 0.9307
## IEAA -0.4774 0.4466 -1.3527 0.3979 0.2851
## EEAA -0.4778 0.6074 -1.6684 0.7128 0.4315
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim <= 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
GEE (Mix, mutual adjust)
In this section, we perform the generalized estimating equations
(GEE) to evaluate the associations between each cluster within current
pollutant exposures and each Epigenetic Age Acceleration with the
formula: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * BC\_PAH6 + \beta_2 PAH31 + \beta_3 NkF
+ \beta_4 PM\_RET + \beta_5 NO2 + \beta_6 SO2 \\
& + \beta_7 * county + \beta_8 * BMI + \beta_9 * ses + \beta_10 *
edu + \epsilon
\end{aligned}
\] where \(Y\) is one of the
epigenetic age accelerations.
Results:
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## The estimated effects:
## $CUR6_BC_PAH6
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -1.4737 0.9206 -3.2780 0.3307 0.1094
## AgeAccelerationResidualHannum -0.9270 0.6848 -2.2692 0.4152 0.1758
## AgeAccelPheno -0.9852 0.6082 -2.1773 0.2069 0.1053
## DNAmAgeSkinBloodClockAdjAge -0.9189 0.6919 -2.2751 0.4372 0.1841
## AgeAccelGrim 0.4061 0.4412 -0.4587 1.2709 0.3574
## DNAmTLAdjAge 0.0279 0.0305 -0.0318 0.0876 0.3592
## IEAA -0.9452 0.6561 -2.2312 0.3407 0.1497
## EEAA -1.5190 0.8958 -3.2747 0.2367 0.0899
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUR6_PAH31
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 1.1989 0.8218 -0.4119 2.8096 0.1446
## AgeAccelerationResidualHannum 0.7096 0.7499 -0.7601 2.1793 0.3440
## AgeAccelPheno 1.5803 0.7041 0.2002 2.9603 0.0248
## DNAmAgeSkinBloodClockAdjAge 1.3248 0.6372 0.0759 2.5736 0.0376
## AgeAccelGrim 0.5440 0.4338 -0.3061 1.3942 0.2097
## DNAmTLAdjAge -0.0276 0.0273 -0.0811 0.0260 0.3131
## IEAA 0.6350 0.7860 -0.9056 2.1755 0.4192
## EEAA 0.9887 0.9497 -0.8727 2.8501 0.2979
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno <= 0.05
## DNAmAgeSkinBloodClockAdjAge <= 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUR6_NkF
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.7286 0.6667 -2.0352 0.5781 0.2745
## AgeAccelerationResidualHannum -0.2519 0.5948 -1.4178 0.9139 0.6719
## AgeAccelPheno -0.6084 0.4813 -1.5517 0.3350 0.2062
## DNAmAgeSkinBloodClockAdjAge -0.1911 0.5542 -1.2774 0.8952 0.7303
## AgeAccelGrim 0.2605 0.3510 -0.4275 0.9485 0.4581
## DNAmTLAdjAge -0.0258 0.0193 -0.0636 0.0120 0.1805
## IEAA -0.5414 0.4452 -1.4139 0.3312 0.2239
## EEAA -0.3847 0.7444 -1.8438 1.0743 0.6053
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUR6_PM_RET
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.0725 0.6160 -1.2800 1.1349 0.9063
## AgeAccelerationResidualHannum -0.6829 0.5457 -1.7525 0.3866 0.2107
## AgeAccelPheno -1.0891 0.6160 -2.2965 0.1184 0.0771
## DNAmAgeSkinBloodClockAdjAge -1.1383 0.5841 -2.2831 0.0064 0.0513
## AgeAccelGrim -0.0872 0.4119 -0.8946 0.7201 0.8323
## DNAmTLAdjAge 0.0233 0.0255 -0.0267 0.0733 0.3608
## IEAA 0.2121 0.6030 -0.9697 1.3940 0.7250
## EEAA -0.8583 0.7149 -2.2595 0.5430 0.2299
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUR6_NO2
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.5896 0.4827 -0.3565 1.5356 0.2219
## AgeAccelerationResidualHannum 0.0024 0.4451 -0.8701 0.8749 0.9957
## AgeAccelPheno 0.3660 0.5295 -0.6717 1.4038 0.4894
## DNAmAgeSkinBloodClockAdjAge 0.5382 0.3939 -0.2339 1.3103 0.1719
## AgeAccelGrim -0.1729 0.3181 -0.7964 0.4505 0.5867
## DNAmTLAdjAge 0.0234 0.0190 -0.0138 0.0607 0.2180
## IEAA 0.4154 0.4241 -0.4157 1.2466 0.3273
## EEAA -0.0089 0.5870 -1.1595 1.1417 0.9879
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUR6_SO2
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.7385 0.6447 -2.0020 0.5250 0.2520
## AgeAccelerationResidualHannum -0.7789 0.5943 -1.9437 0.3859 0.1900
## AgeAccelPheno -1.1245 0.5424 -2.1876 -0.0614 0.0381
## DNAmAgeSkinBloodClockAdjAge -0.9798 0.5251 -2.0091 0.0494 0.0621
## AgeAccelGrim -0.3753 0.2898 -0.9434 0.1928 0.1954
## DNAmTLAdjAge 0.0156 0.0216 -0.0266 0.0579 0.4686
## IEAA -0.7230 0.5661 -1.8325 0.3865 0.2015
## EEAA -1.1491 0.6975 -2.5161 0.2180 0.0995
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno <= 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
Sensitivity analyses
Likelihood ratio (LR) test (no confounders)
Full model: \[
\begin{aligned}
Y = & \beta_0 \\
& + \beta_1 * \text{BC_PAH6} + \beta_2 * \text{PAH31} + \beta_3 *
\text{NkF} + \beta_4 * \text{PM_RET} + \beta_5 * \text{NO2} + \beta_6 *
\text{SO2}\\
& + \epsilon
\end{aligned}
\] Nested model: \[Y = \beta_0 +
\epsilon\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.1940 0.3880
## Hannum EAA 0.4983 0.5293
## PhenoAge EAA 0.0552 0.2208
## Skin&Blood EAA 0.0429 0.2208
## GrimAge EAA 0.1384 0.3691
## DNAmTL 0.5293 0.5293
## IEAA 0.3615 0.4820
## EEAA 0.2863 0.4581
Likelihood ratio (LR) test (no confounders) with subjects using only
smoky or smokeless coal
Full model: \[
\begin{aligned}
Y = & \beta_0 \\
& + \beta_1 * \text{BC_PAH6} + \beta_2 * \text{PAH31} + \beta_3 *
\text{NkF} + \beta_4 * \text{PM_RET} + \beta_5 * \text{NO2} + \beta_6 *
\text{SO2}\\
& + \epsilon
\end{aligned}
\] Nested model: \[Y = \beta_0 +
\epsilon\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.4180 0.4777
## Hannum EAA 0.1468 0.2682
## PhenoAge EAA 0.0227 0.1816
## Skin&Blood EAA 0.1676 0.2682
## GrimAge EAA 0.0801 0.2136
## DNAmTL 0.2540 0.3387
## IEAA 0.6022 0.6022
## EEAA 0.0586 0.2136
Likelihood ratio (LR) test (no confounders) with subjects only using
smoky coal
Full model: \[
\begin{aligned}
Y = & \beta_0 \\
& + \beta_1 * \text{BC_PAH6} + \beta_2 * \text{PAH31} + \beta_3 *
\text{NkF} + \beta_4 * \text{PM_RET} + \beta_5 * \text{NO2} + \beta_6 *
\text{SO2}\\
& + \epsilon
\end{aligned}
\] Nested model: \[Y = \beta_0 +
\epsilon\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.0656 0.0750
## Hannum EAA 0.0030 0.0120
## PhenoAge EAA 0.0107 0.0285
## Skin&Blood EAA 0.0416 0.0750
## GrimAge EAA 0.0647 0.0750
## DNAmTL 0.2183 0.2183
## IEAA 0.0484 0.0750
## EEAA 0.0015 0.0120
GEE (No confounders)
In this section, we perform the generalized estimating equations
(GEE) to evaluate the associations between each cluster within current
pollutant exposures and each Epigenetic Age Acceleration with the
formula: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * X + \epsilon
\end{aligned}
\] where \(Y\) is one of the
epigenetic age accelerations and X is one of the cluster estimates.
Results:
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## The estimated effects:
## $CUR6_BC_PAH6
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.4396 0.5345 -1.4872 0.6081 0.4109
## AgeAccelerationResidualHannum -0.2580 0.3996 -1.0413 0.5252 0.5185
## AgeAccelPheno 0.0509 0.4304 -0.7927 0.8945 0.9058
## DNAmAgeSkinBloodClockAdjAge -0.1448 0.3542 -0.8391 0.5494 0.6826
## AgeAccelGrim 0.5122 0.2570 0.0084 1.0160 0.0463
## DNAmTLAdjAge 0.0227 0.0185 -0.0135 0.0590 0.2186
## IEAA -0.2102 0.4619 -1.1156 0.6952 0.6491
## EEAA -0.5111 0.4941 -1.4795 0.4573 0.3009
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim <= 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUR6_PAH31
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.3145 0.5218 -0.7083 1.3373 0.5467
## AgeAccelerationResidualHannum -0.1046 0.4359 -0.9589 0.7497 0.8104
## AgeAccelPheno 0.3448 0.4782 -0.5924 1.2820 0.4708
## DNAmAgeSkinBloodClockAdjAge 0.2551 0.3535 -0.4378 0.9479 0.4706
## AgeAccelGrim 0.7856 0.2480 0.2996 1.2717 0.0015
## DNAmTLAdjAge -0.0081 0.0165 -0.0403 0.0242 0.6246
## IEAA 0.2237 0.5316 -0.8182 1.2657 0.6739
## EEAA -0.1931 0.5304 -1.2327 0.8466 0.7159
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim <= 0.01
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUR6_NkF
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.0060 0.4487 -0.8855 0.8736 0.9894
## AgeAccelerationResidualHannum -0.1703 0.3897 -0.9341 0.5935 0.6620
## AgeAccelPheno -0.4435 0.3869 -1.2018 0.3147 0.2516
## DNAmAgeSkinBloodClockAdjAge -0.0936 0.3238 -0.7281 0.5410 0.7726
## AgeAccelGrim 0.2990 0.2516 -0.1942 0.7922 0.2347
## DNAmTLAdjAge -0.0283 0.0162 -0.0601 0.0035 0.0811
## IEAA -0.0708 0.3596 -0.7757 0.6341 0.8439
## EEAA -0.1679 0.5151 -1.1775 0.8417 0.7444
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUR6_PM_RET
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.2446 0.4092 -0.5575 1.0467 0.5501
## AgeAccelerationResidualHannum -0.4462 0.3719 -1.1751 0.2827 0.2302
## AgeAccelPheno -0.3131 0.5385 -1.3685 0.7424 0.5610
## DNAmAgeSkinBloodClockAdjAge -0.4074 0.4180 -1.2266 0.4118 0.3297
## AgeAccelGrim 0.5359 0.3358 -0.1222 1.1940 0.1105
## DNAmTLAdjAge 0.0016 0.0198 -0.0371 0.0404 0.9343
## IEAA 0.4289 0.3790 -0.3140 1.1718 0.2578
## EEAA -0.6026 0.5104 -1.6030 0.3977 0.2377
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUR6_NO2
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.4110 0.3834 -0.3405 1.1625 0.2837
## AgeAccelerationResidualHannum -0.1603 0.3554 -0.8570 0.5364 0.6520
## AgeAccelPheno 0.2724 0.4317 -0.5738 1.1185 0.5281
## DNAmAgeSkinBloodClockAdjAge 0.2290 0.2794 -0.3185 0.7766 0.4124
## AgeAccelGrim 0.0075 0.3039 -0.5882 0.6031 0.9804
## DNAmTLAdjAge 0.0108 0.0163 -0.0211 0.0426 0.5079
## IEAA 0.3776 0.3666 -0.3410 1.0962 0.3031
## EEAA -0.2774 0.4852 -1.2285 0.6736 0.5675
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUR6_SO2
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.3262 0.4740 -1.2553 0.6029 0.4913
## AgeAccelerationResidualHannum -0.2930 0.4921 -1.2574 0.6714 0.5515
## AgeAccelPheno -0.5203 0.5847 -1.6663 0.6257 0.3735
## DNAmAgeSkinBloodClockAdjAge -0.3625 0.4188 -1.1834 0.4585 0.3868
## AgeAccelGrim -0.4774 0.2869 -1.0398 0.0850 0.0962
## DNAmTLAdjAge -0.0057 0.0182 -0.0414 0.0300 0.7556
## IEAA -0.4681 0.4600 -1.3697 0.4336 0.3089
## EEAA -0.4333 0.5659 -1.5425 0.6759 0.4439
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
GEE (No confounders, mutual adjust)
In this section, we perform the generalized estimating equations
(GEE) to evaluate the associations between each cluster within current
pollutant exposures and each Epigenetic Age Acceleration with the
formula: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * BC\_PAH6 + \beta_2 PAH31 + \beta_3 NkF
+ \beta_4 PM\_RET + \beta_5 NO2 + \beta_6 SO2 \\
& + \epsilon
\end{aligned}
\] where \(Y\) is one of the
epigenetic age accelerations.
Results:
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## The estimated effects:
## $CUR6_BC_PAH6
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -1.5027 0.9467 -3.3582 0.3528 0.1124
## AgeAccelerationResidualHannum -0.9039 0.6494 -2.1767 0.3690 0.1640
## AgeAccelPheno -1.0664 0.6256 -2.2926 0.1598 0.0883
## DNAmAgeSkinBloodClockAdjAge -0.9810 0.6840 -2.3217 0.3597 0.1515
## AgeAccelGrim 0.2155 0.4685 -0.7027 1.1337 0.6455
## DNAmTLAdjAge 0.0325 0.0287 -0.0238 0.0888 0.2579
## IEAA -0.9745 0.6933 -2.3335 0.3844 0.1599
## EEAA -1.4906 0.8365 -3.1302 0.1489 0.0748
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUR6_PAH31
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 1.2535 0.8323 -0.3779 2.8848 0.1321
## AgeAccelerationResidualHannum 0.7478 0.7111 -0.6459 2.1415 0.2930
## AgeAccelPheno 1.5098 0.7000 0.1378 2.8818 0.0310
## DNAmAgeSkinBloodClockAdjAge 1.3110 0.6359 0.0646 2.5573 0.0392
## AgeAccelGrim 0.4510 0.4652 -0.4608 1.3629 0.3323
## DNAmTLAdjAge -0.0249 0.0273 -0.0784 0.0285 0.3602
## IEAA 0.6306 0.7935 -0.9247 2.1858 0.4268
## EEAA 1.0742 0.8978 -0.6854 2.8338 0.2315
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno <= 0.05
## DNAmAgeSkinBloodClockAdjAge <= 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUR6_NkF
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.5894 0.6960 -1.9535 0.7748 0.3971
## AgeAccelerationResidualHannum -0.2702 0.5732 -1.3936 0.8532 0.6374
## AgeAccelPheno -0.7113 0.4819 -1.6557 0.2331 0.1399
## DNAmAgeSkinBloodClockAdjAge -0.2213 0.5518 -1.3028 0.8602 0.6883
## AgeAccelGrim 0.1906 0.3893 -0.5725 0.9537 0.6244
## DNAmTLAdjAge -0.0211 0.0194 -0.0591 0.0170 0.2771
## IEAA -0.4916 0.4720 -1.4167 0.4335 0.2976
## EEAA -0.3652 0.7128 -1.7623 1.0318 0.6084
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUR6_PM_RET
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.1690 0.6407 -1.4248 1.0867 0.7919
## AgeAccelerationResidualHannum -0.6820 0.5574 -1.7745 0.4105 0.2211
## AgeAccelPheno -0.9336 0.5926 -2.0951 0.2279 0.1151
## DNAmAgeSkinBloodClockAdjAge -1.1297 0.5799 -2.2663 0.0068 0.0514
## AgeAccelGrim 0.1523 0.4124 -0.6560 0.9605 0.7119
## DNAmTLAdjAge 0.0160 0.0249 -0.0327 0.0647 0.5201
## IEAA 0.2398 0.6095 -0.9548 1.4344 0.6940
## EEAA -0.9133 0.7266 -2.3375 0.5108 0.2087
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUR6_NO2
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.4004 0.4516 -0.4848 1.2856 0.3753
## AgeAccelerationResidualHannum 0.0328 0.4317 -0.8133 0.8789 0.9394
## AgeAccelPheno 0.5882 0.4851 -0.3627 1.5390 0.2253
## DNAmAgeSkinBloodClockAdjAge 0.5715 0.3662 -0.1463 1.2892 0.1186
## AgeAccelGrim 0.0380 0.3248 -0.5986 0.6745 0.9069
## DNAmTLAdjAge 0.0132 0.0180 -0.0222 0.0485 0.4660
## IEAA 0.3475 0.4173 -0.4704 1.1654 0.4050
## EEAA -0.0381 0.5511 -1.1183 1.0421 0.9449
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUR6_SO2
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.9251 0.6431 -2.1855 0.3354 0.1503
## AgeAccelerationResidualHannum -0.7003 0.5825 -1.8421 0.4414 0.2292
## AgeAccelPheno -0.9749 0.5515 -2.0558 0.1061 0.0771
## DNAmAgeSkinBloodClockAdjAge -0.9567 0.5289 -1.9933 0.0800 0.0705
## AgeAccelGrim -0.3026 0.2907 -0.8724 0.2672 0.2979
## DNAmTLAdjAge 0.0086 0.0224 -0.0353 0.0524 0.7024
## IEAA -0.8380 0.5622 -1.9400 0.2639 0.1361
## EEAA -1.0844 0.6843 -2.4257 0.2569 0.1131
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
3.2. Clusters based on model-based exposure estimates accrued before
age 18 (clusCHLD5)
The file “LEX_clusCHLD5.csv” has information on estimated pollutant
exposures during early childhood. Estimates are available for 5
different prototypes (cluster variables) for a total of 161 subjects and
211 visits. The prototypes are labelled as:
CHLD5_X7 – a cluster of 7 air pollutants
CHLD5_X33 – a large cluster of 33 air pollutants
CHLD5_NkF – NkF only
CHLD5_NO2 – NO2 only
CHLD5_SO2 – SO2 only
Summary the exposure estimates:
| Characteristic |
N = 129 |
| CHLD5_X7 |
0.11 (-0.5, 0.5) |
| (Missing) |
3 |
| CHLD5_X33 |
0.00 (-0.6, 1.1) |
| (Missing) |
3 |
| CHLD5_NkF |
-0.24 (-0.8, 0.7) |
| (Missing) |
3 |
| CHLD5_NO2 |
0.21 (-0.5, 0.7) |
| (Missing) |
3 |
| CHLD5_SO2 |
0.33 (-1.0, 0.4) |
| (Missing) |
3 |
## By current fuel type:
| Characteristic |
Overall, N = 112 |
Smokeles, N = 17 |
Smoky, N = 87 |
Wood_and_or_Plant, N = 8 |
| CHLD5_X7 |
0.09 (-0.5, 0.5) |
-0.63 (-0.9, -0.1) |
0.10 (-0.5, 0.3) |
0.86 (0.7, 1.1) |
| (Missing) |
3 |
2 |
1 |
0 |
| CHLD5_X33 |
0.23 (-0.7, 1.1) |
-0.83 (-1.4, 0.1) |
0.51 (-0.4, 1.2) |
0.95 (-0.1, 1.0) |
| (Missing) |
3 |
2 |
1 |
0 |
| CHLD5_NkF |
-0.21 (-0.8, 0.7) |
0.06 (-0.3, 0.7) |
-0.45 (-1.0, 0.5) |
1.07 (0.5, 1.5) |
| (Missing) |
3 |
2 |
1 |
0 |
| CHLD5_NO2 |
0.34 (-0.5, 0.8) |
0.17 (-0.5, 0.9) |
0.43 (-0.6, 0.8) |
-0.21 (-0.3, 0.2) |
| (Missing) |
3 |
2 |
1 |
0 |
| CHLD5_SO2 |
0.34 (-0.7, 0.4) |
0.45 (0.3, 1.4) |
0.34 (-0.9, 0.4) |
0.22 (-0.2, 0.3) |
| (Missing) |
3 |
2 |
1 |
0 |
Primary analysis
Likelihood ratio (LR) test (mix model)
Full model: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * \text{X7} + \beta_2 * \text{X33} +
\beta_3 * \text{NkF} + \beta_4 * \text{NO2} + \beta_5 * \text{SO2}\\
& + \beta_6 * county + \beta_7 * BMI + \beta_8 * ses + \beta_{9} *
edu + \epsilon
\end{aligned}
\] Nested model: \[
\begin{aligned}
Y = & \beta_0 \\
& + \beta_1 * county + \beta_2 * BMI + \beta_3 * ses + \beta_4 *
edu + \epsilon
\end{aligned}
\] \(H_0\): The full model and
the nested model fit the data equally well. Thus, you should use the
nested model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.7902 0.7902
## Hannum EAA 0.4596 0.7710
## PhenoAge EAA 0.2823 0.7528
## Skin&Blood EAA 0.2096 0.7528
## GrimAge EAA 0.0066 0.0528
## DNAmTL 0.7835 0.7902
## IEAA 0.6187 0.7902
## EEAA 0.4819 0.7710
GEE (Mix)
In this section, we perform the generalized estimating equations
(GEE) to evaluate the associations between each cluster within current
pollutant exposures and each Epigenetic Age Acceleration with the
formula: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * X \\
& + \beta_2 * county + \beta_3 * BMI + \beta_4 * ses + \beta_5 *
edu + \epsilon
\end{aligned}
\] where \(Y\) is one of the
epigenetic age accelerations and X is one of the cluster estimates.
Results:
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## The estimated effects:
## $CHLD5_X7
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.3328 0.4935 -0.6345 1.3001 0.5001
## AgeAccelerationResidualHannum 0.2637 0.4528 -0.6238 1.1512 0.5603
## AgeAccelPheno 0.5901 0.4893 -0.3690 1.5492 0.2278
## DNAmAgeSkinBloodClockAdjAge 0.1253 0.4167 -0.6915 0.9421 0.7636
## AgeAccelGrim 0.7524 0.2428 0.2766 1.2282 0.0019
## DNAmTLAdjAge 0.0024 0.0179 -0.0327 0.0375 0.8948
## IEAA 0.2947 0.4579 -0.6028 1.1922 0.5199
## EEAA 0.4732 0.5728 -0.6495 1.5959 0.4088
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim <= 0.01
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CHLD5_X33
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.1751 0.4357 -0.6788 1.0291 0.6877
## AgeAccelerationResidualHannum 0.4233 0.4514 -0.4614 1.3080 0.3484
## AgeAccelPheno 1.0052 0.4145 0.1928 1.8176 0.0153
## DNAmAgeSkinBloodClockAdjAge 0.6370 0.3423 -0.0339 1.3080 0.0628
## AgeAccelGrim 0.9202 0.2671 0.3966 1.4438 0.0006
## DNAmTLAdjAge 0.0017 0.0167 -0.0311 0.0345 0.9178
## IEAA -0.1736 0.4286 -1.0136 0.6663 0.6854
## EEAA 0.6738 0.5461 -0.3965 1.7441 0.2172
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno <= 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim <= 0.001
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CHLD5_NkF
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.1088 0.3108 -0.5004 0.7180 0.7263
## AgeAccelerationResidualHannum 0.1589 0.2796 -0.3891 0.7069 0.5698
## AgeAccelPheno -0.1006 0.3453 -0.7773 0.5761 0.7708
## DNAmAgeSkinBloodClockAdjAge -0.1530 0.2998 -0.7407 0.4347 0.6098
## AgeAccelGrim 0.3441 0.2453 -0.1366 0.8249 0.1606
## DNAmTLAdjAge -0.0143 0.0189 -0.0514 0.0228 0.4492
## IEAA -0.0039 0.2819 -0.5564 0.5486 0.9890
## EEAA 0.2225 0.3820 -0.5263 0.9713 0.5603
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CHLD5_NO2
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.0742 0.3383 -0.7374 0.5889 0.8264
## AgeAccelerationResidualHannum 0.1620 0.3611 -0.5457 0.8696 0.6537
## AgeAccelPheno 0.1014 0.3865 -0.6562 0.8590 0.7931
## DNAmAgeSkinBloodClockAdjAge 0.4688 0.2745 -0.0692 1.0068 0.0876
## AgeAccelGrim -0.1935 0.2516 -0.6866 0.2996 0.4419
## DNAmTLAdjAge 0.0035 0.0168 -0.0295 0.0364 0.8370
## IEAA -0.1071 0.2931 -0.6815 0.4673 0.7148
## EEAA 0.1659 0.4583 -0.7324 1.0642 0.7174
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CHLD5_SO2
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.0506 0.5840 -1.0939 1.1952 0.9309
## AgeAccelerationResidualHannum 0.2753 0.5242 -0.7522 1.3028 0.5995
## AgeAccelPheno -0.0021 0.5678 -1.1150 1.1109 0.9971
## DNAmAgeSkinBloodClockAdjAge 0.3932 0.4476 -0.4841 1.2704 0.3797
## AgeAccelGrim -0.2606 0.3112 -0.8706 0.3494 0.4024
## DNAmTLAdjAge 0.0169 0.0215 -0.0252 0.0591 0.4314
## IEAA -0.0827 0.5057 -1.0739 0.9086 0.8702
## EEAA 0.2011 0.6701 -1.1123 1.5144 0.7641
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
GEE (Mix, mutual adjust)
In this section, we perform the generalized estimating equations
(GEE) to evaluate the associations between each cluster within current
pollutant exposures and each Epigenetic Age Acceleration with the
formula: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * X7 + \beta_2 X33 + \beta_3 NkF +
\beta_4 NO2 + \beta_5 SO2 \\
& + \beta_6 * county + \beta_7 * BMI + \beta_8 * ses + \beta_9 *
edu + \epsilon
\end{aligned}
\] where \(Y\) is one of the
epigenetic age accelerations.
Results:
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## The estimated effects:
## $CHLD5_X7
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.5375 0.8931 -1.2128 2.2879 0.5472
## AgeAccelerationResidualHannum 0.1518 0.7423 -1.3032 1.6067 0.8380
## AgeAccelPheno -0.0683 0.8290 -1.6931 1.5564 0.9343
## DNAmAgeSkinBloodClockAdjAge -0.2038 0.7377 -1.6496 1.2421 0.7824
## AgeAccelGrim -0.0784 0.4170 -0.8957 0.7388 0.8508
## DNAmTLAdjAge 0.0299 0.0316 -0.0320 0.0919 0.3436
## IEAA 0.9570 0.7586 -0.5299 2.4439 0.2071
## EEAA 0.1483 0.9876 -1.7874 2.0840 0.8807
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CHLD5_X33
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.1380 0.7662 -1.6397 1.3637 0.8571
## AgeAccelerationResidualHannum 0.3045 0.5969 -0.8656 1.4745 0.6100
## AgeAccelPheno 1.0626 0.6752 -0.2607 2.3859 0.1155
## DNAmAgeSkinBloodClockAdjAge 0.7032 0.6049 -0.4824 1.8888 0.2450
## AgeAccelGrim 1.0012 0.3620 0.2917 1.7107 0.0057
## DNAmTLAdjAge -0.0164 0.0267 -0.0687 0.0359 0.5378
## IEAA -0.7277 0.6455 -1.9929 0.5374 0.2596
## EEAA 0.5607 0.7685 -0.9456 2.0670 0.4657
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim <= 0.01
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CHLD5_NkF
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.0063 0.3320 -0.6570 0.6445 0.9849
## AgeAccelerationResidualHannum 0.1633 0.3060 -0.4365 0.7631 0.5937
## AgeAccelPheno -0.1277 0.3559 -0.8253 0.5699 0.7197
## DNAmAgeSkinBloodClockAdjAge -0.0446 0.2848 -0.6028 0.5136 0.8755
## AgeAccelGrim 0.2820 0.2503 -0.2085 0.7725 0.2599
## DNAmTLAdjAge -0.0192 0.0207 -0.0597 0.0214 0.3543
## IEAA -0.2063 0.2997 -0.7937 0.3811 0.4913
## EEAA 0.2126 0.4238 -0.6181 1.0432 0.6160
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CHLD5_NO2
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.1139 0.3990 -0.8960 0.6682 0.7753
## AgeAccelerationResidualHannum 0.0468 0.3577 -0.6544 0.7479 0.8960
## AgeAccelPheno -0.0169 0.4297 -0.8591 0.8253 0.9686
## DNAmAgeSkinBloodClockAdjAge 0.3188 0.2924 -0.2543 0.8919 0.2756
## AgeAccelGrim -0.2023 0.2516 -0.6955 0.2909 0.4215
## DNAmTLAdjAge -0.0052 0.0197 -0.0438 0.0334 0.7926
## IEAA -0.0469 0.3722 -0.7764 0.6825 0.8997
## EEAA 0.0635 0.4602 -0.8386 0.9655 0.8903
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CHLD5_SO2
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.3605 0.7571 -1.1235 1.8444 0.6340
## AgeAccelerationResidualHannum 0.3247 0.6999 -1.0471 1.6965 0.6427
## AgeAccelPheno -0.1531 0.6274 -1.3829 1.0767 0.8072
## DNAmAgeSkinBloodClockAdjAge 0.0504 0.5068 -0.9429 1.0438 0.9207
## AgeAccelGrim -0.2171 0.3720 -0.9463 0.5120 0.5595
## DNAmTLAdjAge 0.0305 0.0295 -0.0274 0.0883 0.3025
## IEAA 0.3832 0.6912 -0.9715 1.7380 0.5793
## EEAA 0.2247 0.8796 -1.4994 1.9487 0.7984
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
Sensitivity analyses
Likelihood ratio (LR) test (no confounders)
Full model: \[Y = \beta_0 + \beta_1 *
\text{X7} + \beta_2 * \text{X33} + \beta_3 * \text{NkF} + \beta_4 *
\text{NO2} + \beta_5 * \text{SO2} + \epsilon\]
Nested model: \[Y = \beta_0 +
\epsilon\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.9021 0.9021
## Hannum EAA 0.6447 0.8443
## PhenoAge EAA 0.3311 0.8443
## Skin&Blood EAA 0.3499 0.8443
## GrimAge EAA 0.0113 0.0904
## DNAmTL 0.7321 0.8443
## IEAA 0.6910 0.8443
## EEAA 0.7388 0.8443
Likelihood ratio (LR) test (no confounders) with subjects using only
smoky or smokeless coal
Full model: \[Y = \beta_0 + \beta_1 *
\text{X7} + \beta_2 * \text{X33} + \beta_3 * \text{NkF} + \beta_4 *
\text{NO2} + \beta_5 * \text{SO2} + \epsilon\]
Nested model: \[Y = \beta_0 +
\epsilon\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.9832 0.9832
## Hannum EAA 0.3659 0.7010
## PhenoAge EAA 0.0863 0.3452
## Skin&Blood EAA 0.1781 0.4749
## GrimAge EAA 0.0609 0.3452
## DNAmTL 0.5614 0.7485
## IEAA 0.7580 0.8663
## EEAA 0.4381 0.7010
Likelihood ratio (LR) test (no confounders) with subjects only using
smoky coal
Full model: \[Y = \beta_0 + \beta_1 *
\text{X7} + \beta_2 * \text{X33} + \beta_3 * \text{NkF} + \beta_4 *
\text{NO2} + \beta_5 * \text{SO2} + \epsilon\]
Nested model: \[Y = \beta_0 +
\epsilon\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.9319 0.9319
## Hannum EAA 0.3742 0.5987
## PhenoAge EAA 0.0770 0.3100
## Skin&Blood EAA 0.2302 0.5987
## GrimAge EAA 0.0775 0.3100
## DNAmTL 0.3692 0.5987
## IEAA 0.8016 0.9161
## EEAA 0.4628 0.6171
GEE (No confounders)
In this section, we perform the generalized estimating equations
(GEE) to evaluate the associations between each cluster within current
pollutant exposures and each Epigenetic Age Acceleration with the
formula: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * X \\
& + \epsilon
\end{aligned}
\] where \(Y\) is one of the
epigenetic age accelerations and X is one of the cluster estimates.
Results:
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## The estimated effects:
## $CHLD5_X7
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.2293 0.5142 -0.7785 1.2371 0.6556
## AgeAccelerationResidualHannum 0.2397 0.4451 -0.6328 1.1121 0.5903
## AgeAccelPheno 0.6099 0.4913 -0.3530 1.5728 0.2144
## DNAmAgeSkinBloodClockAdjAge 0.1010 0.4283 -0.7385 0.9406 0.8136
## AgeAccelGrim 0.7497 0.2581 0.2438 1.2555 0.0037
## DNAmTLAdjAge 0.0029 0.0177 -0.0318 0.0376 0.8697
## IEAA 0.2553 0.4845 -0.6943 1.2049 0.5982
## EEAA 0.3994 0.5642 -0.7064 1.5053 0.4790
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim <= 0.01
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CHLD5_X33
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.0507 0.4348 -0.8016 0.9029 0.9072
## AgeAccelerationResidualHannum 0.3471 0.4259 -0.4876 1.1818 0.4150
## AgeAccelPheno 0.9328 0.3808 0.1864 1.6792 0.0143
## DNAmAgeSkinBloodClockAdjAge 0.5351 0.3289 -0.1095 1.1796 0.1037
## AgeAccelGrim 0.8261 0.2839 0.2697 1.3824 0.0036
## DNAmTLAdjAge 0.0051 0.0156 -0.0256 0.0357 0.7466
## IEAA -0.1730 0.4338 -1.0233 0.6774 0.6901
## EEAA 0.5122 0.5165 -0.5001 1.5246 0.3213
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno <= 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim <= 0.01
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CHLD5_NkF
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.0968 0.3240 -0.5383 0.7319 0.7652
## AgeAccelerationResidualHannum 0.1682 0.2946 -0.4092 0.7455 0.5681
## AgeAccelPheno -0.0094 0.3524 -0.7002 0.6813 0.9786
## DNAmAgeSkinBloodClockAdjAge -0.1291 0.2767 -0.6714 0.4132 0.6407
## AgeAccelGrim 0.4118 0.2605 -0.0988 0.9223 0.1139
## DNAmTLAdjAge -0.0185 0.0168 -0.0515 0.0144 0.2699
## IEAA -0.0035 0.2869 -0.5658 0.5589 0.9904
## EEAA 0.2252 0.3959 -0.5508 1.0013 0.5695
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CHLD5_NO2
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.2672 0.3047 -0.8644 0.3300 0.3806
## AgeAccelerationResidualHannum 0.0791 0.3128 -0.5339 0.6922 0.8002
## AgeAccelPheno 0.2578 0.3457 -0.4198 0.9354 0.4559
## DNAmAgeSkinBloodClockAdjAge 0.4077 0.2391 -0.0609 0.8763 0.0881
## AgeAccelGrim -0.1213 0.2123 -0.5374 0.2948 0.5679
## DNAmTLAdjAge -0.0006 0.0154 -0.0308 0.0297 0.9705
## IEAA -0.1234 0.2624 -0.6378 0.3909 0.6381
## EEAA -0.0305 0.4107 -0.8355 0.7745 0.9408
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CHLD5_SO2
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.2661 0.5452 -1.3346 0.8025 0.6255
## AgeAccelerationResidualHannum 0.2676 0.4355 -0.5859 1.1211 0.5389
## AgeAccelPheno 0.1891 0.5148 -0.8200 1.1981 0.7135
## DNAmAgeSkinBloodClockAdjAge 0.3531 0.3786 -0.3889 1.0952 0.3509
## AgeAccelGrim -0.2850 0.2825 -0.8387 0.2687 0.3130
## DNAmTLAdjAge 0.0104 0.0191 -0.0271 0.0478 0.5876
## IEAA -0.2151 0.4888 -1.1732 0.7429 0.6598
## EEAA 0.1109 0.5560 -0.9789 1.2007 0.8419
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
GEE (No confounders, mutual adjust)
In this section, we perform the generalized estimating equations
(GEE) to evaluate the associations between each cluster within current
pollutant exposures and each Epigenetic Age Acceleration with the
formula: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * X7 + \beta_2 X33 + \beta_3 NkF +
\beta_4 NO2 + \beta_5 SO2 \\
& + \epsilon
\end{aligned}
\] where \(Y\) is one of the
epigenetic age accelerations.
Results:
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## The estimated effects:
## $CHLD5_X7
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.1542 0.8714 -1.5538 1.8623 0.8595
## AgeAccelerationResidualHannum 0.1198 0.6817 -1.2162 1.4559 0.8605
## AgeAccelPheno 0.1160 0.7743 -1.4017 1.6337 0.8809
## DNAmAgeSkinBloodClockAdjAge -0.2369 0.6866 -1.5826 1.1088 0.7300
## AgeAccelGrim -0.1234 0.3872 -0.8824 0.6357 0.7501
## DNAmTLAdjAge 0.0233 0.0308 -0.0370 0.0835 0.4491
## IEAA 0.7315 0.7440 -0.7267 2.1897 0.3255
## EEAA 0.0345 0.9132 -1.7554 1.8245 0.9699
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CHLD5_X33
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.0098 0.7533 -1.4667 1.4863 0.9896
## AgeAccelerationResidualHannum 0.2661 0.5870 -0.8844 1.4167 0.6503
## AgeAccelPheno 0.8392 0.6226 -0.3811 2.0595 0.1777
## DNAmAgeSkinBloodClockAdjAge 0.6166 0.5482 -0.4579 1.6910 0.2607
## AgeAccelGrim 0.9502 0.3713 0.2224 1.6779 0.0105
## DNAmTLAdjAge -0.0095 0.0260 -0.0604 0.0414 0.7153
## IEAA -0.5891 0.6541 -1.8712 0.6929 0.3678
## EEAA 0.5094 0.7420 -0.9450 1.9638 0.4924
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim <= 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CHLD5_NkF
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.0302 0.3404 -0.6370 0.6974 0.9292
## AgeAccelerationResidualHannum 0.1769 0.3099 -0.4305 0.7844 0.5681
## AgeAccelPheno 0.0038 0.3681 -0.7176 0.7252 0.9918
## DNAmAgeSkinBloodClockAdjAge -0.0150 0.2735 -0.5511 0.5212 0.9564
## AgeAccelGrim 0.4253 0.2618 -0.0877 0.9384 0.1042
## DNAmTLAdjAge -0.0233 0.0184 -0.0594 0.0128 0.2062
## IEAA -0.1897 0.3008 -0.7793 0.3999 0.5283
## EEAA 0.2388 0.4339 -0.6117 1.0892 0.5821
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CHLD5_NO2
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.2186 0.3798 -0.9630 0.5257 0.5648
## AgeAccelerationResidualHannum -0.0820 0.3754 -0.8178 0.6538 0.8271
## AgeAccelPheno 0.1281 0.4045 -0.6646 0.9209 0.7514
## DNAmAgeSkinBloodClockAdjAge 0.2933 0.2964 -0.2876 0.8741 0.3224
## AgeAccelGrim -0.0826 0.2296 -0.5326 0.3674 0.7190
## DNAmTLAdjAge -0.0086 0.0192 -0.0462 0.0291 0.6552
## IEAA -0.0313 0.3454 -0.7082 0.6457 0.9278
## EEAA -0.1588 0.4849 -1.1092 0.7916 0.7433
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CHLD5_SO2
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.0692 0.7041 -1.4493 1.3109 0.9217
## AgeAccelerationResidualHannum 0.3577 0.6176 -0.8528 1.5682 0.5624
## AgeAccelPheno 0.0517 0.5713 -1.0679 1.1714 0.9278
## DNAmAgeSkinBloodClockAdjAge 0.0078 0.4361 -0.8470 0.8626 0.9857
## AgeAccelGrim -0.3245 0.3228 -0.9573 0.3083 0.3149
## DNAmTLAdjAge 0.0217 0.0274 -0.0320 0.0754 0.4290
## IEAA 0.1010 0.6351 -1.1439 1.3458 0.8737
## EEAA 0.2001 0.7612 -1.2919 1.6920 0.7927
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
3.3. Clusters based on model-based lifetime exposure estimates
(clusCUM6)
The file “LEX_clus CUM6.csv” has information on estimated cumulative
pollutant exposures during the lifecourse. Estimates are available for 6
different prototypes (cluster variables) for a total of 161 subjects and
211 visits. The prototypes are labelled as:
CUM6_BC_NO2_PM – a cluster of BC, NO2, and PM
CUM6_PAH36 – a large cluster of 36 PAHs
CUM6_DlP – DlP only
CUM6_NkF – NkF only
CUM6_RET – retene only
CUM6_SO2 – SO2 only
Summary the exposure estimates:
| Characteristic |
N = 129 |
| CUM6_BC_NO2_PM |
0.28 (-0.6, 0.9) |
| (Missing) |
3 |
| CUM6_PAH36 |
0.25 (-0.6, 1.1) |
| (Missing) |
3 |
| CUM6_DlP |
-0.47 (-1.0, 0.8) |
| (Missing) |
3 |
| CUM6_NkF |
-0.09 (-0.8, 0.4) |
| (Missing) |
3 |
| CUM6_RET |
-0.22 (-0.7, 0.4) |
| (Missing) |
3 |
| CUM6_SO2 |
0.09 (-0.8, 0.3) |
| (Missing) |
3 |
## By current fuel type:
| Characteristic |
Overall, N = 112 |
Smokeles, N = 17 |
Smoky, N = 87 |
Wood_and_or_Plant, N = 8 |
| CUM6_BC_NO2_PM |
0.22 (-0.6, 0.8) |
0.19 (-0.3, 0.7) |
0.10 (-1.0, 0.8) |
1.38 (0.4, 1.6) |
| (Missing) |
3 |
2 |
1 |
0 |
| CUM6_PAH36 |
0.25 (-0.6, 1.1) |
-1.00 (-1.2, -0.3) |
0.32 (-0.5, 1.2) |
0.83 (0.4, 1.4) |
| (Missing) |
3 |
2 |
1 |
0 |
| CUM6_DlP |
-0.48 (-1.0, 0.8) |
0.65 (0.5, 1.1) |
-0.66 (-1.2, 0.7) |
0.42 (0.3, 0.6) |
| (Missing) |
3 |
2 |
1 |
0 |
| CUM6_NkF |
-0.22 (-0.8, 0.5) |
-0.07 (-0.3, 0.4) |
-0.31 (-1.0, 0.4) |
1.18 (0.1, 1.7) |
| (Missing) |
3 |
2 |
1 |
0 |
| CUM6_RET |
-0.22 (-0.7, 0.3) |
-0.41 (-0.9, 0.3) |
-0.25 (-0.8, 0.2) |
1.71 (1.2, 1.9) |
| (Missing) |
3 |
2 |
1 |
0 |
| CUM6_SO2 |
0.09 (-0.4, 0.4) |
1.13 (0.5, 1.6) |
-0.03 (-0.9, 0.3) |
-0.02 (-0.6, 0.1) |
| (Missing) |
3 |
2 |
1 |
0 |
Primary analysis
Likelihood ratio (LR) test (mix model)
Full model: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * \text{BC_NO2_PM} + \beta_2 *
\text{PAH36} + \beta_3 * \text{DlP} + \beta_4 * \text{NkF} + \beta_5 *
\text{RET} + \beta_6 * \text{SO2}\\
& + \beta_7 * county + \beta_8 * BMI + \beta_9 * ses + \beta_{10}
* edu + \epsilon
\end{aligned}
\] Nested model: \[
\begin{aligned}
Y = & \beta_0 \\
& + \beta_1 * county + \beta_2 * BMI + \beta_3 * ses + \beta_4 *
edu + \epsilon
\end{aligned}
\] \(H_0\): The full model and
the nested model fit the data equally well. Thus, you should use the
nested model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.4471 0.5961
## Hannum EAA 0.7507 0.7507
## PhenoAge EAA 0.2961 0.4738
## Skin&Blood EAA 0.2805 0.4738
## GrimAge EAA 0.0004 0.0032
## DNAmTL 0.1577 0.4738
## IEAA 0.2939 0.4738
## EEAA 0.6381 0.7293
GEE (Mix)
In this section, we perform the generalized estimating equations
(GEE) to evaluate the associations between each cluster within current
pollutant exposures and each Epigenetic Age Acceleration with the
formula: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * X \\
& + \beta_2 * county + \beta_3 * BMI + \beta_4 * ses + \beta_5 *
edu + \epsilon
\end{aligned}
\] where \(Y\) is one of the
epigenetic age accelerations and X is one of the cluster estimates.
Results:
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## The estimated effects:
## $CUM6_BC_NO2_PM
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.4786 0.5704 -0.6394 1.5967 0.4014
## AgeAccelerationResidualHannum 0.5152 0.5265 -0.5167 1.5472 0.3278
## AgeAccelPheno 0.3777 0.5950 -0.7886 1.5439 0.5256
## DNAmAgeSkinBloodClockAdjAge -0.0071 0.5818 -1.1475 1.1332 0.9902
## AgeAccelGrim 0.9077 0.3443 0.2329 1.5825 0.0084
## DNAmTLAdjAge -0.0240 0.0188 -0.0610 0.0129 0.2024
## IEAA 0.6290 0.5653 -0.4789 1.7369 0.2658
## EEAA 0.8229 0.6469 -0.4451 2.0909 0.2034
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim <= 0.01
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUM6_PAH36
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.4288 0.5009 -0.5530 1.4106 0.3920
## AgeAccelerationResidualHannum 0.3748 0.4646 -0.5357 1.2854 0.4198
## AgeAccelPheno 0.8962 0.4748 -0.0345 1.8268 0.0591
## DNAmAgeSkinBloodClockAdjAge 0.5333 0.3895 -0.2302 1.2967 0.1710
## AgeAccelGrim 1.1373 0.2520 0.6435 1.6312 0.0000
## DNAmTLAdjAge -0.0126 0.0181 -0.0481 0.0228 0.4857
## IEAA 0.3233 0.4790 -0.6156 1.2621 0.4997
## EEAA 0.5687 0.5557 -0.5205 1.6579 0.3061
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim <= 0.001
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUM6_DlP
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.5116 0.4935 -0.4557 1.4789 0.2999
## AgeAccelerationResidualHannum 0.0323 0.4617 -0.8727 0.9372 0.9443
## AgeAccelPheno -0.0624 0.4323 -0.9097 0.7849 0.8852
## DNAmAgeSkinBloodClockAdjAge 0.1294 0.3741 -0.6039 0.8627 0.7295
## AgeAccelGrim -0.4971 0.2268 -0.9417 -0.0525 0.0284
## DNAmTLAdjAge -0.0239 0.0184 -0.0599 0.0121 0.1935
## IEAA 0.5841 0.4381 -0.2747 1.4428 0.1825
## EEAA 0.1843 0.5795 -0.9515 1.3201 0.7505
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim <= 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUM6_NkF
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.2161 0.3560 -0.4816 0.9139 0.5438
## AgeAccelerationResidualHannum 0.1657 0.3699 -0.5594 0.8907 0.6543
## AgeAccelPheno -0.0775 0.4058 -0.8729 0.7179 0.8485
## DNAmAgeSkinBloodClockAdjAge -0.1047 0.3530 -0.7966 0.5871 0.7667
## AgeAccelGrim 0.5782 0.2254 0.1364 1.0200 0.0103
## DNAmTLAdjAge -0.0380 0.0174 -0.0720 -0.0039 0.0291
## IEAA 0.1227 0.3111 -0.4870 0.7324 0.6933
## EEAA 0.2863 0.4888 -0.6718 1.2445 0.5581
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim <= 0.05
## DNAmTLAdjAge <= 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUM6_RET
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.2691 0.4450 -0.6031 1.1414 0.5453
## AgeAccelerationResidualHannum -0.0012 0.3568 -0.7005 0.6981 0.9973
## AgeAccelPheno -0.1489 0.4803 -1.0903 0.7924 0.7565
## DNAmAgeSkinBloodClockAdjAge -0.2516 0.4283 -1.0911 0.5878 0.5568
## AgeAccelGrim 0.4833 0.3252 -0.1542 1.1207 0.1373
## DNAmTLAdjAge -0.0054 0.0176 -0.0399 0.0291 0.7594
## IEAA 0.3880 0.3965 -0.3891 1.1652 0.3277
## EEAA 0.0425 0.4851 -0.9084 0.9933 0.9303
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUM6_SO2
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.1809 0.5663 -1.2908 0.9291 0.7494
## AgeAccelerationResidualHannum -0.0030 0.5637 -1.1078 1.1019 0.9958
## AgeAccelPheno -0.4582 0.5884 -1.6115 0.6951 0.4361
## DNAmAgeSkinBloodClockAdjAge 0.0715 0.4785 -0.8664 1.0095 0.8812
## AgeAccelGrim -0.4746 0.3137 -1.0894 0.1402 0.1302
## DNAmTLAdjAge 0.0229 0.0223 -0.0207 0.0666 0.3028
## IEAA -0.0673 0.4926 -1.0327 0.8982 0.8914
## EEAA -0.2555 0.6815 -1.5912 1.0803 0.7078
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
GEE (Mix, mutual adjust)
In this section, we perform the generalized estimating equations
(GEE) to evaluate the associations between each cluster within current
pollutant exposures and each Epigenetic Age Acceleration with the
formula: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * BC\_NO2\_PM + \beta_2 PAH36 + \beta_3
DlP + \beta_4 NkF + \beta_5 RET + \beta_6 SO2 \\
& + \beta_7 * county + \beta_8 * BMI + \beta_9 * ses + \beta_10 *
edu + \epsilon
\end{aligned}
\] where \(Y\) is one of the
epigenetic age accelerations.
Results:
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## The estimated effects:
## $CUM6_BC_NO2_PM
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.2730 0.6524 -1.5517 1.0057 0.6756
## AgeAccelerationResidualHannum 0.6151 0.7013 -0.7594 1.9896 0.3804
## AgeAccelPheno -0.2835 0.8236 -1.8977 1.3307 0.7306
## DNAmAgeSkinBloodClockAdjAge -0.7631 0.6624 -2.0615 0.5352 0.2493
## AgeAccelGrim 0.7149 0.4833 -0.2323 1.6622 0.1391
## DNAmTLAdjAge -0.0325 0.0281 -0.0876 0.0226 0.2481
## IEAA -0.0894 0.6934 -1.4484 1.2696 0.8974
## EEAA 0.9116 0.8688 -0.7913 2.6144 0.2941
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUM6_PAH36
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.9323 0.6682 -0.3774 2.2420 0.1630
## AgeAccelerationResidualHannum 0.1279 0.6700 -1.1853 1.4410 0.8486
## AgeAccelPheno 1.5842 0.7983 0.0196 3.1489 0.0472
## DNAmAgeSkinBloodClockAdjAge 1.4449 0.6380 0.1944 2.6953 0.0235
## AgeAccelGrim 0.5238 0.4610 -0.3797 1.4273 0.2558
## DNAmTLAdjAge 0.0023 0.0288 -0.0542 0.0589 0.9355
## IEAA 0.7026 0.6603 -0.5916 1.9967 0.2873
## EEAA 0.2849 0.8493 -1.3797 1.9496 0.7373
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno <= 0.05
## DNAmAgeSkinBloodClockAdjAge <= 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUM6_DlP
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 1.1621 0.6226 -0.0581 2.3824 0.0620
## AgeAccelerationResidualHannum -0.1086 0.5988 -1.2823 1.0652 0.8561
## AgeAccelPheno 0.7049 0.5797 -0.4313 1.8410 0.2240
## DNAmAgeSkinBloodClockAdjAge 0.8192 0.5197 -0.1995 1.8378 0.1150
## AgeAccelGrim -0.5787 0.3413 -1.2476 0.0902 0.0899
## DNAmTLAdjAge -0.0074 0.0256 -0.0576 0.0428 0.7720
## IEAA 1.2352 0.5457 0.1657 2.3048 0.0236
## EEAA 0.1156 0.7762 -1.4057 1.6369 0.8816
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA <= 0.05
## EEAA > 0.05
##
## $CUM6_NkF
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.4062 0.4337 -1.2563 0.4439 0.3490
## AgeAccelerationResidualHannum 0.2405 0.3949 -0.5336 1.0146 0.5426
## AgeAccelPheno -0.3634 0.4882 -1.3203 0.5934 0.4566
## DNAmAgeSkinBloodClockAdjAge -0.4152 0.4260 -1.2502 0.4199 0.3298
## AgeAccelGrim 0.6814 0.2909 0.1113 1.2515 0.0191
## DNAmTLAdjAge -0.0433 0.0209 -0.0842 -0.0024 0.0379
## IEAA -0.6019 0.3794 -1.3455 0.1416 0.1126
## EEAA 0.2872 0.5563 -0.8031 1.3775 0.6056
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim <= 0.05
## DNAmTLAdjAge <= 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUM6_RET
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.3670 0.6210 -0.8501 1.5841 0.5545
## AgeAccelerationResidualHannum -0.4159 0.5437 -1.4815 0.6497 0.4443
## AgeAccelPheno -0.4622 0.5574 -1.5547 0.6302 0.4069
## DNAmAgeSkinBloodClockAdjAge -0.1741 0.5141 -1.1818 0.8336 0.7348
## AgeAccelGrim -0.4531 0.4344 -1.3044 0.3983 0.2969
## DNAmTLAdjAge 0.0305 0.0221 -0.0128 0.0739 0.1674
## IEAA 0.6335 0.5167 -0.3791 1.6462 0.2201
## EEAA -0.5941 0.7039 -1.9737 0.7855 0.3986
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUM6_SO2
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.2686 0.5881 -1.4213 0.8841 0.6479
## AgeAccelerationResidualHannum -0.0950 0.6022 -1.2752 1.0852 0.8747
## AgeAccelPheno -0.6224 0.5838 -1.7667 0.5219 0.2864
## DNAmAgeSkinBloodClockAdjAge -0.0324 0.4924 -0.9976 0.9327 0.9475
## AgeAccelGrim -0.3984 0.3069 -0.9999 0.2031 0.1942
## DNAmTLAdjAge 0.0335 0.0231 -0.0119 0.0789 0.1477
## IEAA -0.1197 0.4982 -1.0961 0.8566 0.8100
## EEAA -0.4561 0.7410 -1.9085 0.9962 0.5382
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
Sensitivity analyses
Likelihood ratio (LR) test (no confounders)
Full model: \[Y = \beta_0 + \beta_1 *
\text{BC_NO2_PM} + \beta_2 * \text{PAH36} + \beta_3 * \text{DlP} +
\beta_4 * \text{NkF} + \beta_5 * \text{RET} + \beta_6 * \text{SO2} +
\epsilon\]
Nested model: \[Y = \beta_0 +
\epsilon\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.5999 0.7999
## Hannum EAA 0.9345 0.9345
## PhenoAge EAA 0.2861 0.6744
## Skin&Blood EAA 0.3810 0.6744
## GrimAge EAA 0.0106 0.0848
## DNAmTL 0.1197 0.4788
## IEAA 0.4215 0.6744
## EEAA 0.8973 0.9345
Likelihood ratio (LR) test (no confounders) with subjects using only
smoky or smokeless coal
Full model: \[Y = \beta_0 + \beta_1 *
\text{BC_NO2_PM} + \beta_2 * \text{PAH36} + \beta_3 * \text{DlP} +
\beta_4 * \text{NkF} + \beta_5 * \text{RET} + \beta_6 * \text{SO2} +
\epsilon\]
Nested model: \[Y = \beta_0 +
\epsilon\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.6468 0.6468
## Hannum EAA 0.5227 0.5974
## PhenoAge EAA 0.0258 0.2064
## Skin&Blood EAA 0.1287 0.2574
## GrimAge EAA 0.1020 0.2574
## DNAmTL 0.0800 0.2574
## IEAA 0.4186 0.5974
## EEAA 0.5204 0.5974
Likelihood ratio (LR) test (no confounders) with subjects only using
smoky coal
Full model: \[Y = \beta_0 + \beta_1 *
\text{BC_NO2_PM} + \beta_2 * \text{PAH36} + \beta_3 * \text{DlP} +
\beta_4 * \text{NkF} + \beta_5 * \text{RET} + \beta_6 * \text{SO2} +
\epsilon\]
Nested model: \[Y = \beta_0 +
\epsilon\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.6400 0.6400
## Hannum EAA 0.3301 0.4401
## PhenoAge EAA 0.0252 0.2016
## Skin&Blood EAA 0.0857 0.2018
## GrimAge EAA 0.1009 0.2018
## DNAmTL 0.0841 0.2018
## IEAA 0.4278 0.4889
## EEAA 0.3192 0.4401
GEE (No confounders)
In this section, we perform the generalized estimating equations
(GEE) to evaluate the associations between each cluster within current
pollutant exposures and each Epigenetic Age Acceleration with the
formula: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * X + \epsilon
\end{aligned}
\] where \(Y\) is one of the
epigenetic age accelerations and X is one of the cluster estimates.
Results:
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## The estimated effects:
## $CUM6_BC_NO2_PM
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.1884 0.3968 -0.9660 0.5893 0.6350
## AgeAccelerationResidualHannum 0.3993 0.3004 -0.1895 0.9880 0.1838
## AgeAccelPheno 0.5024 0.4156 -0.3122 1.3170 0.2267
## DNAmAgeSkinBloodClockAdjAge -0.0883 0.3483 -0.7710 0.5943 0.7997
## AgeAccelGrim 0.5445 0.2453 0.0638 1.0252 0.0264
## DNAmTLAdjAge -0.0227 0.0166 -0.0554 0.0099 0.1718
## IEAA 0.1498 0.3613 -0.5583 0.8579 0.6784
## EEAA 0.4604 0.4025 -0.3284 1.2493 0.2526
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim <= 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUM6_PAH36
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.0036 0.4229 -0.8324 0.8252 0.9932
## AgeAccelerationResidualHannum 0.3575 0.3682 -0.3641 1.0791 0.3315
## AgeAccelPheno 0.8447 0.3868 0.0865 1.6029 0.0290
## DNAmAgeSkinBloodClockAdjAge 0.3168 0.3249 -0.3201 0.9537 0.3296
## AgeAccelGrim 0.8579 0.2642 0.3401 1.3757 0.0012
## DNAmTLAdjAge -0.0123 0.0159 -0.0434 0.0188 0.4376
## IEAA 0.0988 0.4034 -0.6919 0.8895 0.8065
## EEAA 0.4204 0.4479 -0.4575 1.2982 0.3480
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno <= 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim <= 0.01
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUM6_DlP
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.2182 0.4332 -0.6308 1.0672 0.6145
## AgeAccelerationResidualHannum 0.0243 0.3848 -0.7300 0.7786 0.9496
## AgeAccelPheno 0.2035 0.3663 -0.5145 0.9214 0.5786
## DNAmAgeSkinBloodClockAdjAge 0.1096 0.2981 -0.4746 0.6938 0.7131
## AgeAccelGrim -0.1918 0.2177 -0.6185 0.2350 0.3785
## DNAmTLAdjAge -0.0277 0.0138 -0.0548 -0.0006 0.0450
## IEAA 0.4099 0.3883 -0.3512 1.1710 0.2912
## EEAA 0.0564 0.4831 -0.8904 1.0032 0.9071
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge <= 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUM6_NkF
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.0429 0.3555 -0.6537 0.7396 0.9039
## AgeAccelerationResidualHannum 0.1869 0.3603 -0.5193 0.8932 0.6039
## AgeAccelPheno 0.1030 0.3965 -0.6740 0.8801 0.7949
## DNAmAgeSkinBloodClockAdjAge -0.1298 0.3189 -0.7549 0.4952 0.6839
## AgeAccelGrim 0.5998 0.2254 0.1580 1.0415 0.0078
## DNAmTLAdjAge -0.0396 0.0149 -0.0688 -0.0104 0.0079
## IEAA 0.0726 0.3010 -0.5174 0.6626 0.8094
## EEAA 0.2414 0.4768 -0.6932 1.1760 0.6127
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim <= 0.01
## DNAmTLAdjAge <= 0.01
## IEAA > 0.05
## EEAA > 0.05
##
## $CUM6_RET
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.0873 0.4603 -0.8148 0.9894 0.8495
## AgeAccelerationResidualHannum 0.0181 0.3516 -0.6711 0.7072 0.9590
## AgeAccelPheno -0.0092 0.4657 -0.9219 0.9035 0.9843
## DNAmAgeSkinBloodClockAdjAge -0.2701 0.4132 -1.0799 0.5398 0.5134
## AgeAccelGrim 0.5156 0.3173 -0.1064 1.1376 0.1042
## DNAmTLAdjAge -0.0093 0.0174 -0.0434 0.0247 0.5909
## IEAA 0.3182 0.3985 -0.4627 1.0992 0.4245
## EEAA -0.0022 0.4759 -0.9350 0.9306 0.9963
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUM6_SO2
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.5367 0.4805 -1.4785 0.4050 0.2639
## AgeAccelerationResidualHannum 0.0624 0.4131 -0.7471 0.8720 0.8798
## AgeAccelPheno -0.0746 0.5017 -1.0579 0.9086 0.8817
## DNAmAgeSkinBloodClockAdjAge 0.0566 0.3702 -0.6689 0.7822 0.8784
## AgeAccelGrim -0.3999 0.2539 -0.8976 0.0978 0.1153
## DNAmTLAdjAge 0.0096 0.0177 -0.0252 0.0443 0.5893
## IEAA -0.2477 0.4356 -1.1015 0.6060 0.5696
## EEAA -0.2363 0.4984 -1.2132 0.7407 0.6355
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
GEE (No confounders, mutual adjust)
In this section, we perform the generalized estimating equations
(GEE) to evaluate the associations between each cluster within current
pollutant exposures and each Epigenetic Age Acceleration with the
formula: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * BC\_NO2\_PM + \beta_2 PAH36 + \beta_3
DlP + \beta_4 NkF + \beta_5 RET + \beta_6 SO2 \\
& + \epsilon
\end{aligned}
\] where \(Y\) is one of the
epigenetic age accelerations.
Results:
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## The estimated effects:
## $CUM6_BC_NO2_PM
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.7127 0.5955 -1.8800 0.4545 0.2314
## AgeAccelerationResidualHannum 0.6054 0.5958 -0.5623 1.7732 0.3095
## AgeAccelPheno -0.2482 0.8394 -1.8934 1.3970 0.7675
## DNAmAgeSkinBloodClockAdjAge -0.8953 0.6236 -2.1175 0.3269 0.1511
## AgeAccelGrim 0.3240 0.4702 -0.5976 1.2455 0.4908
## DNAmTLAdjAge -0.0300 0.0296 -0.0880 0.0281 0.3115
## IEAA -0.4600 0.5869 -1.6104 0.6904 0.4332
## EEAA 0.8186 0.8015 -0.7523 2.3894 0.3071
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUM6_PAH36
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.7815 0.6724 -0.5364 2.0995 0.2451
## AgeAccelerationResidualHannum 0.0064 0.6754 -1.3173 1.3301 0.9925
## AgeAccelPheno 1.6047 0.8088 0.0195 3.1900 0.0472
## DNAmAgeSkinBloodClockAdjAge 1.3850 0.6435 0.1237 2.6464 0.0314
## AgeAccelGrim 0.5224 0.4881 -0.4342 1.4790 0.2845
## DNAmTLAdjAge 0.0043 0.0277 -0.0501 0.0587 0.8768
## IEAA 0.6628 0.6674 -0.6452 1.9709 0.3206
## EEAA 0.0605 0.8465 -1.5987 1.7196 0.9431
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno <= 0.05
## DNAmAgeSkinBloodClockAdjAge <= 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUM6_DlP
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.9648 0.6171 -0.2447 2.1744 0.1179
## AgeAccelerationResidualHannum -0.2146 0.5478 -1.2884 0.8591 0.6952
## AgeAccelPheno 0.9482 0.5944 -0.2169 2.1132 0.1107
## DNAmAgeSkinBloodClockAdjAge 0.8418 0.5075 -0.1530 1.8366 0.0972
## AgeAccelGrim -0.2335 0.3869 -0.9918 0.5248 0.5461
## DNAmTLAdjAge -0.0109 0.0229 -0.0559 0.0340 0.6333
## IEAA 1.1205 0.5518 0.0391 2.2020 0.0423
## EEAA -0.1084 0.7307 -1.5407 1.3238 0.8820
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA <= 0.05
## EEAA > 0.05
##
## $CUM6_NkF
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.3678 0.4506 -1.2509 0.5153 0.4143
## AgeAccelerationResidualHannum 0.2938 0.4300 -0.5489 1.1365 0.4944
## AgeAccelPheno -0.3958 0.4946 -1.3651 0.5735 0.4235
## DNAmAgeSkinBloodClockAdjAge -0.4530 0.4363 -1.3082 0.4022 0.2992
## AgeAccelGrim 0.6049 0.3250 -0.0322 1.2420 0.0627
## DNAmTLAdjAge -0.0426 0.0207 -0.0832 -0.0020 0.0399
## IEAA -0.6206 0.3863 -1.3776 0.1365 0.1081
## EEAA 0.3705 0.5999 -0.8053 1.5462 0.5369
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge <= 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUM6_RET
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.3001 0.6469 -0.9677 1.5680 0.6427
## AgeAccelerationResidualHannum -0.4716 0.5510 -1.5515 0.6084 0.3921
## AgeAccelPheno -0.3517 0.5555 -1.4406 0.7371 0.5267
## DNAmAgeSkinBloodClockAdjAge -0.1332 0.5120 -1.1367 0.8703 0.7947
## AgeAccelGrim -0.2467 0.4382 -1.1057 0.6123 0.5735
## DNAmTLAdjAge 0.0283 0.0216 -0.0140 0.0707 0.1901
## IEAA 0.6105 0.5223 -0.4132 1.6343 0.2424
## EEAA -0.6961 0.7125 -2.0925 0.7004 0.3286
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $CUM6_SO2
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.6641 0.5296 -1.7021 0.3738 0.2098
## AgeAccelerationResidualHannum -0.1502 0.5194 -1.1682 0.8679 0.7725
## AgeAccelPheno -0.4912 0.5618 -1.5924 0.6101 0.3820
## DNAmAgeSkinBloodClockAdjAge -0.0404 0.4557 -0.9336 0.8529 0.9294
## AgeAccelGrim -0.5686 0.2749 -1.1073 -0.0298 0.0386
## DNAmTLAdjAge 0.0325 0.0206 -0.0079 0.0730 0.1151
## IEAA -0.4426 0.4453 -1.3154 0.4302 0.3202
## EEAA -0.6035 0.6288 -1.8360 0.6290 0.3372
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim <= 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
3.4. Clusters based on pollutant measurements (clusMEAS6)
The file “LEX_clusMEAS6.csv” has information on measured pollutant
exposures during each visit. Estimates are available for 6 different
prototypes (cluster variables) for a total of 54 subjects and 54 visits.
The prototypes are labelled as:
MEAS6_BC_ PM_RET – a cluster of BC, PM, and retene
MEAS6_X31 – a large cluster of 31 air pollutants
MEAS6_X5 – a smaller cluster of 5 air pollutants
MEAS6_DlP – DlP only
MEAS6_NkF – NkF only
MEAS6_ NO2_SO2 – NO2, and SO2
Summary the exposure estimates:
| Characteristic |
N = 129 |
| MEAS6_BC_PM_RET |
-0.11 (-0.6, 0.5) |
| (Missing) |
64 |
| MEAS6_X31 |
0.14 (-0.7, 0.7) |
| (Missing) |
64 |
| MEAS6_X5 |
-0.10 (-1.0, 1.0) |
| (Missing) |
64 |
| MEAS6_DlP |
-0.63 (-0.7, 1.3) |
| (Missing) |
64 |
| MEAS6_NkF |
-0.50 (-0.6, 0.9) |
| (Missing) |
64 |
| MEAS6_NO2_SO2 |
-0.04 (-0.8, 0.9) |
| (Missing) |
64 |
## By current fuel type:
| Characteristic |
Overall, N = 112 |
Smokeles, N = 17 |
Smoky, N = 87 |
Wood_and_or_Plant, N = 8 |
| MEAS6_BC_PM_RET |
-0.14 (-0.6, 0.5) |
-0.84 (-2.0, -0.2) |
0.06 (-0.5, 0.6) |
1.08 (0.2, 2.4) |
| (Missing) |
52 |
3 |
48 |
1 |
| MEAS6_X31 |
0.14 (-0.7, 0.8) |
-1.02 (-2.0, -0.8) |
0.29 (0.0, 0.9) |
0.82 (-0.1, 0.9) |
| (Missing) |
52 |
3 |
48 |
1 |
| MEAS6_X5 |
-0.22 (-1.1, 1.0) |
-1.06 (-1.1, -1.0) |
0.65 (-0.8, 1.2) |
0.55 (-0.3, 1.1) |
| (Missing) |
52 |
3 |
48 |
1 |
| MEAS6_DlP |
-0.62 (-0.7, 1.3) |
0.53 (-0.5, 1.2) |
-0.69 (-0.7, 1.5) |
-0.30 (-0.6, 0.5) |
| (Missing) |
52 |
3 |
48 |
1 |
| MEAS6_NkF |
-0.50 (-0.6, 0.4) |
-0.46 (-0.6, -0.4) |
-0.51 (-0.6, 0.4) |
0.16 (-0.6, 1.0) |
| (Missing) |
52 |
3 |
48 |
1 |
| MEAS6_NO2_SO2 |
0.11 (-0.8, 1.0) |
1.28 (0.7, 1.8) |
-0.43 (-0.9, 0.8) |
0.20 (-0.8, 0.6) |
| (Missing) |
52 |
3 |
48 |
1 |
Primary analysis
Likelihood ratio (LR) test (mix model)
Full model: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * \text{BC_PM_RET} + \beta_2 * \text{X31}
+ \beta_3 * \text{X5} + \beta_4 * \text{DlP} + \beta_5 * \text{NkF} +
\beta_6 * \text{NO2_SO2}\\
& + \beta_7 * county + \beta_8 * BMI + \beta_9 * ses + \beta_{10}
* edu + \epsilon
\end{aligned}
\] Nested model: \[
\begin{aligned}
Y = & \beta_0 \\
& + \beta_1 * county + \beta_2 * BMI + \beta_3 * ses + \beta_4 *
edu + \epsilon
\end{aligned}
\] \(H_0\): The full model and
the nested model fit the data equally well. Thus, you should use the
nested model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.0715 0.1144
## Hannum EAA 0.0283 0.0755
## PhenoAge EAA 0.1941 0.2588
## Skin&Blood EAA 0.0058 0.0464
## GrimAge EAA 0.0547 0.1094
## DNAmTL 0.5550 0.5550
## IEAA 0.3445 0.3937
## EEAA 0.0187 0.0748
GEE (Mix)
In this section, we perform the generalized estimating equations
(GEE) to evaluate the associations between each cluster within current
pollutant exposures and each Epigenetic Age Acceleration with the
formula: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * X \\
& + \beta_2 * county + \beta_3 * BMI + \beta_4 * ses + \beta_5 *
edu + \epsilon
\end{aligned}
\] where \(Y\) is one of the
epigenetic age accelerations and X is one of the cluster estimates.
Results:
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## The estimated effects:
## $MEAS6_BC_PM_RET
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.2131 0.4434 -1.0821 0.6558 0.6307
## AgeAccelerationResidualHannum -0.3690 0.4019 -1.1567 0.4188 0.3586
## AgeAccelPheno 0.2287 0.7026 -1.1483 1.6058 0.7447
## DNAmAgeSkinBloodClockAdjAge -0.0268 0.3873 -0.7859 0.7324 0.9449
## AgeAccelGrim 0.6926 0.4739 -0.2362 1.6214 0.1439
## DNAmTLAdjAge -0.0015 0.0295 -0.0594 0.0564 0.9606
## IEAA -0.4851 0.3891 -1.2478 0.2776 0.2126
## EEAA -0.4596 0.4796 -1.3996 0.4804 0.3379
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $MEAS6_X31
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 1.0372 0.5907 -0.1207 2.1950 0.0791
## AgeAccelerationResidualHannum 0.7603 0.5751 -0.3669 1.8876 0.1862
## AgeAccelPheno 1.3384 0.6067 0.1493 2.5275 0.0274
## DNAmAgeSkinBloodClockAdjAge 1.2287 0.5097 0.2297 2.2276 0.0159
## AgeAccelGrim 1.1050 0.3110 0.4953 1.7146 0.0004
## DNAmTLAdjAge -0.0265 0.0240 -0.0735 0.0205 0.2690
## IEAA 0.4088 0.4659 -0.5043 1.3220 0.3802
## EEAA 0.9734 0.6286 -0.2587 2.2055 0.1215
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno <= 0.05
## DNAmAgeSkinBloodClockAdjAge <= 0.05
## AgeAccelGrim <= 0.001
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $MEAS6_X5
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.1163 0.6791 -1.4475 1.2148 0.8640
## AgeAccelerationResidualHannum -0.6886 0.7466 -2.1520 0.7747 0.3564
## AgeAccelPheno 0.5990 0.7070 -0.7866 1.9847 0.3968
## DNAmAgeSkinBloodClockAdjAge 0.6521 0.5849 -0.4942 1.7985 0.2648
## AgeAccelGrim 1.0605 0.3884 0.2992 1.8218 0.0063
## DNAmTLAdjAge 0.0259 0.0289 -0.0308 0.0826 0.3707
## IEAA 0.1573 0.5446 -0.9100 1.2247 0.7726
## EEAA -1.0902 0.8962 -2.8467 0.6663 0.2238
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim <= 0.01
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $MEAS6_DlP
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.3307 0.7517 -1.1427 1.8041 0.6600
## AgeAccelerationResidualHannum 0.1404 0.6613 -1.1558 1.4366 0.8319
## AgeAccelPheno 0.3856 0.7079 -1.0019 1.7731 0.5859
## DNAmAgeSkinBloodClockAdjAge -0.8663 0.5664 -1.9765 0.2440 0.1262
## AgeAccelGrim -0.2806 0.5080 -1.2763 0.7150 0.5806
## DNAmTLAdjAge -0.0294 0.0330 -0.0941 0.0354 0.3738
## IEAA 0.3886 0.5756 -0.7396 1.5168 0.4996
## EEAA 0.1581 0.8367 -1.4817 1.7980 0.8501
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $MEAS6_NkF
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.1099 0.6676 -1.1987 1.4185 0.8693
## AgeAccelerationResidualHannum 0.6275 0.5196 -0.3909 1.6458 0.2272
## AgeAccelPheno 0.2914 0.7025 -1.0855 1.6683 0.6783
## DNAmAgeSkinBloodClockAdjAge 0.2283 0.5030 -0.7576 1.2142 0.6499
## AgeAccelGrim -0.1056 0.3829 -0.8560 0.6448 0.7826
## DNAmTLAdjAge -0.0091 0.0259 -0.0598 0.0417 0.7259
## IEAA -0.2750 0.5928 -1.4369 0.8870 0.6428
## EEAA 0.9232 0.7028 -0.4542 2.3007 0.1890
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $MEAS6_NO2_SO2
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.1663 0.5685 -1.2806 0.9479 0.7698
## AgeAccelerationResidualHannum 0.6345 0.5420 -0.4278 1.6967 0.2417
## AgeAccelPheno -0.0485 0.5306 -1.0885 0.9915 0.9271
## DNAmAgeSkinBloodClockAdjAge -0.1171 0.3961 -0.8934 0.6593 0.7676
## AgeAccelGrim -0.3701 0.4232 -1.1996 0.4593 0.3818
## DNAmTLAdjAge -0.0228 0.0338 -0.0890 0.0433 0.4988
## IEAA -0.2743 0.4718 -1.1990 0.6505 0.5611
## EEAA 0.3185 0.6518 -0.9591 1.5960 0.6251
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
GEE (Mix, mutual adjust)
In this section, we perform the generalized estimating equations
(GEE) to evaluate the associations between each cluster within current
pollutant exposures and each Epigenetic Age Acceleration with the
formula: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * BC\_PAH6 + \beta_2 PAH31 + \beta_3 NkF
+ \beta_4 PM\_RET + \beta_5 NO2 + \beta_6 SO2 \\
& + \beta_7 * county + \beta_8 * BMI + \beta_9 * ses + \beta_10 *
edu + \epsilon
\end{aligned}
\] where \(Y\) is one of the
epigenetic age accelerations.
Results:
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## The estimated effects:
## $MEAS6_BC_PM_RET
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.8052 0.7632 -2.3010 0.6906 0.2914
## AgeAccelerationResidualHannum -0.7006 0.5989 -1.8745 0.4733 0.2421
## AgeAccelPheno -0.6131 0.9323 -2.4405 1.2142 0.5108
## DNAmAgeSkinBloodClockAdjAge -0.9711 0.6184 -2.1831 0.2410 0.1163
## AgeAccelGrim 0.0925 0.6425 -1.1667 1.3518 0.8855
## DNAmTLAdjAge -0.0012 0.0320 -0.0640 0.0617 0.9712
## IEAA -0.9818 0.6638 -2.2829 0.3193 0.1392
## EEAA -0.8869 0.6880 -2.2353 0.4615 0.1973
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $MEAS6_X31
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 2.5915 1.0286 0.5754 4.6075 0.0118
## AgeAccelerationResidualHannum 2.3493 0.6263 1.1217 3.5769 0.0002
## AgeAccelPheno 1.9240 0.9541 0.0540 3.7941 0.0437
## DNAmAgeSkinBloodClockAdjAge 2.3571 0.6772 1.0299 3.6844 0.0005
## AgeAccelGrim 1.1725 0.5775 0.0405 2.3044 0.0423
## DNAmTLAdjAge -0.0840 0.0386 -0.1596 -0.0084 0.0293
## IEAA 1.1612 0.8849 -0.5732 2.8955 0.1894
## EEAA 3.2309 0.8370 1.5904 4.8715 0.0001
## sig_level
## AgeAccelerationResidual <= 0.05
## AgeAccelerationResidualHannum <= 0.001
## AgeAccelPheno <= 0.05
## DNAmAgeSkinBloodClockAdjAge <= 0.001
## AgeAccelGrim <= 0.05
## DNAmTLAdjAge <= 0.05
## IEAA > 0.05
## EEAA <= 0.001
##
## $MEAS6_X5
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -1.7309 1.1968 -4.0767 0.6150 0.1481
## AgeAccelerationResidualHannum -1.9693 1.1237 -4.1717 0.2332 0.0797
## AgeAccelPheno -0.4652 1.0716 -2.5655 1.6352 0.6642
## DNAmAgeSkinBloodClockAdjAge -0.7981 0.8475 -2.4593 0.8630 0.3463
## AgeAccelGrim -0.0180 0.8073 -1.6003 1.5644 0.9822
## DNAmTLAdjAge 0.0877 0.0460 -0.0024 0.1778 0.0564
## IEAA -0.2995 1.0465 -2.3506 1.7516 0.7747
## EEAA -2.9864 1.3631 -5.6580 -0.3147 0.0285
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA <= 0.05
##
## $MEAS6_DlP
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.2058 0.7052 -1.5881 1.1764 0.7704
## AgeAccelerationResidualHannum -0.4345 0.5723 -1.5562 0.6871 0.4477
## AgeAccelPheno 0.1030 0.6246 -1.1212 1.3272 0.8690
## DNAmAgeSkinBloodClockAdjAge -1.2570 0.5106 -2.2577 -0.2562 0.0138
## AgeAccelGrim -0.3899 0.4144 -1.2022 0.4225 0.3469
## DNAmTLAdjAge -0.0106 0.0251 -0.0599 0.0387 0.6741
## IEAA 0.2093 0.6092 -0.9847 1.4033 0.7311
## EEAA -0.6479 0.6782 -1.9773 0.6815 0.3394
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge <= 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $MEAS6_NkF
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.7312 0.7189 -2.1402 0.6778 0.3091
## AgeAccelerationResidualHannum -0.0519 0.6828 -1.3902 1.2864 0.9394
## AgeAccelPheno -0.1264 0.7417 -1.5801 1.3274 0.8647
## DNAmAgeSkinBloodClockAdjAge -0.3063 0.5391 -1.3630 0.7504 0.5699
## AgeAccelGrim -0.4543 0.5249 -1.4831 0.5745 0.3868
## DNAmTLAdjAge 0.0235 0.0309 -0.0372 0.0841 0.4479
## IEAA -0.5339 0.7324 -1.9695 0.9016 0.4660
## EEAA -0.1986 0.8403 -1.8455 1.4484 0.8132
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $MEAS6_NO2_SO2
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.2723 0.5813 -1.4116 0.8670 0.6394
## AgeAccelerationResidualHannum 0.6689 0.5030 -0.3171 1.6549 0.1836
## AgeAccelPheno 0.0393 0.5672 -1.0725 1.1510 0.9448
## DNAmAgeSkinBloodClockAdjAge -0.0496 0.4552 -0.9419 0.8426 0.9132
## AgeAccelGrim -0.2664 0.4393 -1.1274 0.5947 0.5443
## DNAmTLAdjAge -0.0335 0.0332 -0.0985 0.0315 0.3121
## IEAA -0.4687 0.4682 -1.3863 0.4490 0.3168
## EEAA 0.3302 0.5877 -0.8216 1.4821 0.5742
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
Sensitivity analyses
Likelihood ratio (LR) test (no confounders)
Full model: \[Y = \beta_0 + \beta_1 *
\text{BC_PM_RET} + \beta_2 * \text{X31} + \beta_3 * \text{X5} + \beta_4
* \text{DlP} + \beta_5 * \text{NkF} + \beta_6 * \text{NO2_SO2} +
\epsilon\]
Nested model: \[Y = \beta_0 +
\epsilon\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.0310 0.0620
## Hannum EAA 0.0064 0.0256
## PhenoAge EAA 0.1701 0.2268
## Skin&Blood EAA 0.0124 0.0331
## GrimAge EAA 0.0983 0.1573
## DNAmTL 0.2723 0.3112
## IEAA 0.4332 0.4332
## EEAA 0.0027 0.0216
Likelihood ratio (LR) test (no confounders) with subjects using only
smoky or smokeless coal
Full model: \[Y = \beta_0 + \beta_1 *
\text{BC_PAH6} + \beta_2 * \text{PAH31} + \beta_3 * \text{NkF} + \beta_4
* \text{PM_RET} + \beta_5 * \text{NO2} + \beta_6 * \text{SO2} +
\epsilon\]
Nested model: \[Y = \beta_0 +
\epsilon\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.1053 0.1404
## Hannum EAA 0.0718 0.1400
## PhenoAge EAA 0.1504 0.1719
## Skin&Blood EAA 0.0509 0.1400
## GrimAge EAA 0.0193 0.1400
## DNAmTL 0.5776 0.5776
## IEAA 0.0875 0.1400
## EEAA 0.0604 0.1400
Likelihood ratio (LR) test (no confounders) with subjects only using
smoky coal
Full model: \[Y = \beta_0 + \beta_1 *
\text{BC_PM_RET} + \beta_2 * \text{X31} + \beta_3 * \text{X5} + \beta_4
* \text{DlP} + \beta_5 * \text{NkF} + \beta_6 * \text{NO2_SO2} +
\epsilon\]
Nested model: \[Y = \beta_0 +
\epsilon\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.1171 0.1792
## Hannum EAA 0.0625 0.1250
## PhenoAge EAA 0.1344 0.1792
## Skin&Blood EAA 0.0589 0.1250
## GrimAge EAA 0.0083 0.0664
## DNAmTL 0.5076 0.5076
## IEAA 0.5013 0.5076
## EEAA 0.0334 0.1250
GEE (no confounders)
In this section, we perform the generalized estimating equations
(GEE) to evaluate the associations between each cluster within current
pollutant exposures and each Epigenetic Age Acceleration with the
formula: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * X + \epsilon
\end{aligned}
\] where \(Y\) is one of the
epigenetic age accelerations and X is one of the cluster estimates.
Results:
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## The estimated effects:
## $MEAS6_BC_PM_RET
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.2167 0.4383 -1.0758 0.6423 0.6209
## AgeAccelerationResidualHannum -0.4474 0.4061 -1.2434 0.3485 0.2705
## AgeAccelPheno 0.1582 0.6700 -1.1550 1.4713 0.8134
## DNAmAgeSkinBloodClockAdjAge -0.1368 0.3571 -0.8366 0.5630 0.7017
## AgeAccelGrim 0.6923 0.4524 -0.1944 1.5791 0.1260
## DNAmTLAdjAge 0.0073 0.0293 -0.0503 0.0648 0.8047
## IEAA -0.3835 0.3918 -1.1513 0.3844 0.3276
## EEAA -0.6213 0.5111 -1.6231 0.3804 0.2241
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $MEAS6_X31
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 1.0149 0.6577 -0.2742 2.3040 0.1228
## AgeAccelerationResidualHannum 0.6948 0.6320 -0.5440 1.9336 0.2716
## AgeAccelPheno 1.1231 0.5958 -0.0447 2.2909 0.0594
## DNAmAgeSkinBloodClockAdjAge 1.0622 0.5431 -0.0022 2.1266 0.0505
## AgeAccelGrim 1.0435 0.3271 0.4024 1.6847 0.0014
## DNAmTLAdjAge -0.0187 0.0252 -0.0681 0.0307 0.4589
## IEAA 0.4805 0.4937 -0.4872 1.4482 0.3304
## EEAA 0.8324 0.7485 -0.6347 2.2996 0.2661
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim <= 0.01
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $MEAS6_X5
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.3006 0.6067 -1.4897 0.8885 0.6202
## AgeAccelerationResidualHannum -0.8748 0.5914 -2.0339 0.2843 0.1391
## AgeAccelPheno 0.3478 0.5754 -0.7799 1.4755 0.5455
## DNAmAgeSkinBloodClockAdjAge 0.2598 0.5129 -0.7455 1.2651 0.6125
## AgeAccelGrim 0.7729 0.3759 0.0362 1.5096 0.0397
## DNAmTLAdjAge 0.0403 0.0248 -0.0083 0.0888 0.1041
## IEAA 0.2191 0.4639 -0.6901 1.1283 0.6367
## EEAA -1.4382 0.7134 -2.8365 -0.0399 0.0438
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim <= 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA <= 0.05
##
## $MEAS6_DlP
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.4196 0.6529 -0.8601 1.6994 0.5204
## AgeAccelerationResidualHannum 0.2971 0.6100 -0.8984 1.4926 0.6262
## AgeAccelPheno 0.4584 0.6170 -0.7510 1.6678 0.4575
## DNAmAgeSkinBloodClockAdjAge -0.5615 0.5524 -1.6442 0.5212 0.3094
## AgeAccelGrim -0.1811 0.5272 -1.2143 0.8522 0.7312
## DNAmTLAdjAge -0.0384 0.0304 -0.0980 0.0212 0.2066
## IEAA 0.2697 0.4931 -0.6966 1.2361 0.5843
## EEAA 0.4537 0.7754 -1.0662 1.9736 0.5585
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $MEAS6_NkF
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.2503 0.7174 -1.1558 1.6564 0.7272
## AgeAccelerationResidualHannum 0.7301 0.5467 -0.3414 1.8017 0.1817
## AgeAccelPheno 0.2821 0.6799 -1.0505 1.6147 0.6782
## DNAmAgeSkinBloodClockAdjAge 0.2450 0.5915 -0.9144 1.4044 0.6787
## AgeAccelGrim 0.0144 0.4106 -0.7904 0.8192 0.9720
## DNAmTLAdjAge -0.0127 0.0251 -0.0619 0.0364 0.6120
## IEAA -0.2325 0.6420 -1.4908 1.0258 0.7172
## EEAA 1.0794 0.7402 -0.3713 2.5301 0.1447
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $MEAS6_NO2_SO2
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.2145 0.6073 -0.9757 1.4047 0.7239
## AgeAccelerationResidualHannum 0.7399 0.5173 -0.2741 1.7538 0.1527
## AgeAccelPheno 0.1041 0.5513 -0.9765 1.1847 0.8502
## DNAmAgeSkinBloodClockAdjAge 0.3653 0.4951 -0.6050 1.3357 0.4606
## AgeAccelGrim -0.0571 0.3785 -0.7991 0.6848 0.8801
## DNAmTLAdjAge -0.0276 0.0268 -0.0802 0.0249 0.3022
## IEAA 0.1130 0.5503 -0.9656 1.1916 0.8373
## EEAA 0.5420 0.6373 -0.7071 1.7911 0.3950
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
GEE (Mix, mutual adjust)
In this section, we perform the generalized estimating equations
(GEE) to evaluate the associations between each cluster within current
pollutant exposures and each Epigenetic Age Acceleration with the
formula: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * BC\_PAH6 + \beta_2 PAH31 + \beta_3 NkF
+ \beta_4 PM\_RET + \beta_5 NO2 + \beta_6 SO2 \\
& + \beta_7 * county + \beta_8 * BMI + \beta_9 * ses + \beta_10 *
edu + \epsilon
\end{aligned}
\] where \(Y\) is one of the
epigenetic age accelerations.
Results:
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## [1] "Fitting with 65 observations."
## The estimated effects:
## $MEAS6_BC_PM_RET
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.7407 0.7753 -2.2603 0.7790 0.3394
## AgeAccelerationResidualHannum -0.6831 0.6184 -1.8951 0.5289 0.2693
## AgeAccelPheno -0.6034 0.9533 -2.4718 1.2651 0.5268
## DNAmAgeSkinBloodClockAdjAge -0.8952 0.5949 -2.0611 0.2708 0.1324
## AgeAccelGrim 0.1771 0.6602 -1.1170 1.4712 0.7885
## DNAmTLAdjAge -0.0018 0.0347 -0.0699 0.0663 0.9594
## IEAA -0.9163 0.6674 -2.2245 0.3919 0.1698
## EEAA -0.8543 0.7222 -2.2699 0.5613 0.2369
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $MEAS6_X31
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 2.5894 1.0393 0.5523 4.6265 0.0127
## AgeAccelerationResidualHannum 2.2527 0.6493 0.9801 3.5253 0.0005
## AgeAccelPheno 1.6932 0.9645 -0.1972 3.5836 0.0792
## DNAmAgeSkinBloodClockAdjAge 2.2402 0.6968 0.8744 3.6059 0.0013
## AgeAccelGrim 1.1844 0.5342 0.1375 2.2313 0.0266
## DNAmTLAdjAge -0.0791 0.0338 -0.1454 -0.0129 0.0192
## IEAA 1.1619 0.9394 -0.6793 3.0031 0.2161
## EEAA 3.1071 0.8293 1.4817 4.7326 0.0002
## sig_level
## AgeAccelerationResidual <= 0.05
## AgeAccelerationResidualHannum <= 0.001
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge <= 0.01
## AgeAccelGrim <= 0.05
## DNAmTLAdjAge <= 0.05
## IEAA > 0.05
## EEAA <= 0.001
##
## $MEAS6_X5
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -1.7267 1.1445 -3.9699 0.5165 0.1314
## AgeAccelerationResidualHannum -1.9205 0.9412 -3.7652 -0.0758 0.0413
## AgeAccelPheno -0.3795 0.9556 -2.2525 1.4935 0.6913
## DNAmAgeSkinBloodClockAdjAge -0.9976 0.8396 -2.6432 0.6479 0.2347
## AgeAccelGrim -0.2187 0.8025 -1.7916 1.3542 0.7852
## DNAmTLAdjAge 0.0897 0.0389 0.0135 0.1659 0.0210
## IEAA -0.1290 1.0229 -2.1339 1.8758 0.8996
## EEAA -3.0352 1.1266 -5.2433 -0.8271 0.0071
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum <= 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge <= 0.05
## IEAA > 0.05
## EEAA <= 0.01
##
## $MEAS6_DlP
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.1338 0.6936 -1.4932 1.2256 0.8470
## AgeAccelerationResidualHannum -0.3661 0.5512 -1.4464 0.7143 0.5066
## AgeAccelPheno 0.3079 0.5690 -0.8074 1.4232 0.5885
## DNAmAgeSkinBloodClockAdjAge -0.9609 0.5164 -1.9730 0.0512 0.0628
## AgeAccelGrim -0.2218 0.4382 -1.0806 0.6370 0.6128
## DNAmTLAdjAge -0.0129 0.0233 -0.0585 0.0327 0.5796
## IEAA 0.1340 0.5884 -1.0194 1.2873 0.8199
## EEAA -0.4883 0.6544 -1.7710 0.7943 0.4555
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $MEAS6_NkF
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.6478 0.7652 -2.1477 0.8520 0.3972
## AgeAccelerationResidualHannum -0.0370 0.6745 -1.3589 1.2850 0.9563
## AgeAccelPheno -0.1091 0.7730 -1.6242 1.4060 0.8877
## DNAmAgeSkinBloodClockAdjAge -0.3303 0.6485 -1.6013 0.9407 0.6105
## AgeAccelGrim -0.3798 0.5342 -1.4268 0.6673 0.4772
## DNAmTLAdjAge 0.0231 0.0297 -0.0352 0.0814 0.4374
## IEAA -0.4115 0.7783 -1.9368 1.1139 0.5970
## EEAA -0.1914 0.8294 -1.8171 1.4343 0.8175
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $MEAS6_NO2_SO2
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.0374 0.6678 -1.2714 1.3462 0.9553
## AgeAccelerationResidualHannum 0.6406 0.5799 -0.4959 1.7772 0.2693
## AgeAccelPheno 0.1016 0.5779 -1.0312 1.2343 0.8605
## DNAmAgeSkinBloodClockAdjAge 0.3942 0.5205 -0.6260 1.4144 0.4488
## AgeAccelGrim 0.0744 0.4134 -0.7357 0.8846 0.8571
## DNAmTLAdjAge -0.0204 0.0264 -0.0721 0.0313 0.4388
## IEAA -0.0215 0.5930 -1.1837 1.1407 0.9711
## EEAA 0.3338 0.6838 -1.0064 1.6740 0.6255
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
3.5. Clusters based on urinary biomarkers (clusURI5)
The file “LEX_clusURI5.csv” has information on measured urinary
biomarkers obtained during each visit. Estimates are available for 5
different prototypes (cluster variables) for a total of 163 subjects and
186 visits. The prototypes are labelled as:
URI5_NAP_1M_2M – a cluster of Naphthalene, 1Methylnaphthalene, and
2Methylnaphthalene
URI5_ACE – Acenaphthene only
URI5_FLU_PHE – Fluoranthene and Phenanthrene_anth
URI5_PYR – Pyrene only
URI5_CHR – Baa_Chrysene only
Summary the exposure estimates:
| Characteristic |
N = 129 |
| MEAS6_BC_PM_RET |
-0.11 (-0.6, 0.5) |
| (Missing) |
64 |
| MEAS6_X31 |
0.14 (-0.7, 0.7) |
| (Missing) |
64 |
| MEAS6_X5 |
-0.10 (-1.0, 1.0) |
| (Missing) |
64 |
| MEAS6_DlP |
-0.63 (-0.7, 1.3) |
| (Missing) |
64 |
| MEAS6_NkF |
-0.50 (-0.6, 0.9) |
| (Missing) |
64 |
| MEAS6_NO2_SO2 |
-0.04 (-0.8, 0.9) |
| (Missing) |
64 |
## By current fuel type:
| Characteristic |
Overall, N = 112 |
Smokeles, N = 17 |
Smoky, N = 87 |
Wood_and_or_Plant, N = 8 |
| MEAS6_BC_PM_RET |
-0.14 (-0.6, 0.5) |
-0.84 (-2.0, -0.2) |
0.06 (-0.5, 0.6) |
1.08 (0.2, 2.4) |
| (Missing) |
52 |
3 |
48 |
1 |
| MEAS6_X31 |
0.14 (-0.7, 0.8) |
-1.02 (-2.0, -0.8) |
0.29 (0.0, 0.9) |
0.82 (-0.1, 0.9) |
| (Missing) |
52 |
3 |
48 |
1 |
| MEAS6_X5 |
-0.22 (-1.1, 1.0) |
-1.06 (-1.1, -1.0) |
0.65 (-0.8, 1.2) |
0.55 (-0.3, 1.1) |
| (Missing) |
52 |
3 |
48 |
1 |
| MEAS6_DlP |
-0.62 (-0.7, 1.3) |
0.53 (-0.5, 1.2) |
-0.69 (-0.7, 1.5) |
-0.30 (-0.6, 0.5) |
| (Missing) |
52 |
3 |
48 |
1 |
| MEAS6_NkF |
-0.50 (-0.6, 0.4) |
-0.46 (-0.6, -0.4) |
-0.51 (-0.6, 0.4) |
0.16 (-0.6, 1.0) |
| (Missing) |
52 |
3 |
48 |
1 |
| MEAS6_NO2_SO2 |
0.11 (-0.8, 1.0) |
1.28 (0.7, 1.8) |
-0.43 (-0.9, 0.8) |
0.20 (-0.8, 0.6) |
| (Missing) |
52 |
3 |
48 |
1 |
Primary analysis
Likelihood ratio (LR) test (mix model)
Full model: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * \text{NAP_1M_2M} + \beta_2 * \text{ACE}
+ \beta_3 * \text{FLU_PHE} + \beta_4 * \text{PYR} + \beta_5 *
\text{CHR}\\
& + \beta_6 * county + \beta_7 * BMI + \beta_8 * ses + \beta_{9} *
edu + \epsilon
\end{aligned}
\] Nested model: \[
\begin{aligned}
Y = & \beta_0 \\
& + \beta_1 * county + \beta_2 * BMI + \beta_3 * ses + \beta_4 *
edu + \epsilon
\end{aligned}
\] \(H_0\): The full model and
the nested model fit the data equally well. Thus, you should use the
nested model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.7671 0.8767
## Hannum EAA 0.5794 0.8528
## PhenoAge EAA 0.0267 0.2136
## Skin&Blood EAA 0.3915 0.7830
## GrimAge EAA 0.0782 0.3128
## DNAmTL 0.1529 0.4077
## IEAA 0.8826 0.8826
## EEAA 0.6396 0.8528
GEE (Mix)
In this section, we perform the generalized estimating equations
(GEE) to evaluate the associations between each cluster within current
pollutant exposures and each Epigenetic Age Acceleration with the
formula: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * X \\
& + \beta_2 * county + \beta_3 * BMI + \beta_4 * ses + \beta_5 *
edu + \epsilon
\end{aligned}
\] where \(Y\) is one of the
epigenetic age accelerations and X is one of the cluster estimates.
Results:
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## The estimated effects:
## $URI5_NAP_1M_2M
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.1207 0.4892 -0.8380 1.0795 0.8051
## AgeAccelerationResidualHannum -0.5311 0.4726 -1.4574 0.3951 0.2610
## AgeAccelPheno -0.0765 0.4506 -0.9597 0.8067 0.8652
## DNAmAgeSkinBloodClockAdjAge -0.2610 0.3845 -1.0147 0.4927 0.4973
## AgeAccelGrim 0.4613 11.5825 -22.2404 23.1629 0.9682
## DNAmTLAdjAge -0.0190 0.0274 -0.0726 0.0346 0.4875
## IEAA -0.0709 0.5298 -1.1093 0.9675 0.8936
## EEAA -0.0672 0.5250 -1.0963 0.9618 0.8981
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $URI5_ACE
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.0718 0.4161 -0.8873 0.7437 0.8630
## AgeAccelerationResidualHannum 0.4692 0.3255 -0.1687 1.1071 0.1494
## AgeAccelPheno 0.8548 0.4681 -0.0628 1.7723 0.0679
## DNAmAgeSkinBloodClockAdjAge -0.0606 0.3144 -0.6768 0.5557 0.8473
## AgeAccelGrim -1.0904 0.3341 -1.7452 -0.4356 0.0011
## DNAmTLAdjAge -0.0036 0.0164 -0.0358 0.0285 0.8254
## IEAA 0.2453 0.4113 -0.5609 1.0515 0.5509
## EEAA 0.5891 0.4421 -0.2773 1.4556 0.1827
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim <= 0.01
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $URI5_FLU_PHE
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.2247 0.4454 -0.6483 1.0977 0.6139
## AgeAccelerationResidualHannum 0.2299 0.4475 -0.6472 1.1070 0.6074
## AgeAccelPheno 0.3113 0.4054 -0.4832 1.1058 0.4425
## DNAmAgeSkinBloodClockAdjAge 0.0430 0.3263 -0.5964 0.6825 0.8950
## AgeAccelGrim 0.7761 0.7236 -0.6421 2.1943 0.2835
## DNAmTLAdjAge -0.0294 0.0419 -0.1114 0.0527 0.4831
## IEAA 0.1451 0.4563 -0.7491 1.0394 0.7504
## EEAA 0.5353 0.5835 -0.6083 1.6789 0.3589
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $URI5_PYR
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.0171 0.4598 -0.9183 0.8841 0.9703
## AgeAccelerationResidualHannum 0.4174 0.4215 -0.4088 1.2436 0.3221
## AgeAccelPheno 1.0257 0.4853 0.0745 1.9770 0.0346
## DNAmAgeSkinBloodClockAdjAge 0.6843 0.4160 -0.1311 1.4998 0.1000
## AgeAccelGrim 0.6208 0.3417 -0.0490 1.2905 0.0693
## DNAmTLAdjAge -0.0339 0.0265 -0.0859 0.0181 0.2018
## IEAA -0.0710 0.4712 -0.9945 0.8525 0.8802
## EEAA 0.6476 0.5407 -0.4122 1.7073 0.2310
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno <= 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $URI5_CHR
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.0824 0.3638 -0.6306 0.7954 0.8208
## AgeAccelerationResidualHannum 0.1025 0.4291 -0.7384 0.9435 0.8111
## AgeAccelPheno -0.0586 0.3702 -0.7843 0.6671 0.8743
## DNAmAgeSkinBloodClockAdjAge 0.0933 0.3141 -0.5222 0.7089 0.7663
## AgeAccelGrim 0.3337 0.3381 -0.3289 0.9964 0.3236
## DNAmTLAdjAge -0.0195 0.2864 -0.5808 0.5419 0.9458
## IEAA -0.2279 0.4158 -1.0428 0.5870 0.5836
## EEAA 0.3170 0.5072 -0.6771 1.3111 0.5320
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
GEE (Mix, mutual adjust)
In this section, we perform the generalized estimating equations
(GEE) to evaluate the associations between each cluster within current
pollutant exposures and each Epigenetic Age Acceleration with the
formula: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * NAP\_1M\_2M + \beta_2 ACE + \beta_3
FLU\_PHE + \beta_4 PYR + \beta_5 CHR \\
& + \beta_6 * county + \beta_7 * BMI + \beta_8 * ses + \beta_9 *
edu + \epsilon
\end{aligned}
\] where \(Y\) is one of the
epigenetic age accelerations.
Results:
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## The estimated effects:
## $URI5_NAP_1M_2M
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.0437 0.5992 -1.1308 1.2182 0.9418
## AgeAccelerationResidualHannum -1.2381 0.5210 -2.2593 -0.2169 0.0175
## AgeAccelPheno -0.4501 0.5919 -1.6102 0.7101 0.4470
## DNAmAgeSkinBloodClockAdjAge -0.4391 0.5182 -1.4548 0.5765 0.3968
## AgeAccelGrim 0.0399 2.3489 -4.5640 4.6438 0.9864
## DNAmTLAdjAge -0.0189 0.0901 -0.1955 0.1576 0.8336
## IEAA -0.3374 0.6410 -1.5937 0.9188 0.5986
## EEAA -0.6963 0.6453 -1.9611 0.5685 0.2806
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum <= 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $URI5_ACE
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.0682 0.4130 -0.8776 0.7413 0.8689
## AgeAccelerationResidualHannum 0.5431 0.3021 -0.0491 1.1353 0.0723
## AgeAccelPheno 0.8730 0.4527 -0.0143 1.7602 0.0538
## DNAmAgeSkinBloodClockAdjAge -0.0531 0.2924 -0.6262 0.5199 0.8558
## AgeAccelGrim 0.3481 1.2322 -2.0671 2.7633 0.7776
## DNAmTLAdjAge -0.0326 0.2517 -0.5259 0.4607 0.8969
## IEAA 0.2118 0.4206 -0.6125 1.0361 0.6145
## EEAA 0.6167 0.4508 -0.2669 1.5003 0.1714
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $URI5_FLU_PHE
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.2405 0.7019 -1.1353 1.6162 0.7319
## AgeAccelerationResidualHannum 1.1556 0.5839 0.0111 2.3001 0.0478
## AgeAccelPheno 0.0397 0.6082 -1.1525 1.2318 0.9480
## DNAmAgeSkinBloodClockAdjAge -0.1676 0.4686 -1.0860 0.7508 0.7206
## AgeAccelGrim 0.3150 3.3020 -6.1569 6.7869 0.9240
## DNAmTLAdjAge -0.0296 0.1850 -0.3921 0.3330 0.8730
## IEAA 0.6990 0.6426 -0.5604 1.9585 0.2766
## EEAA 0.6951 0.7493 -0.7736 2.1637 0.3536
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum <= 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $URI5_PYR
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.1845 0.5705 -1.3028 0.9338 0.7464
## AgeAccelerationResidualHannum 0.1875 0.3734 -0.5444 0.9193 0.6156
## AgeAccelPheno 1.1701 0.5287 0.1339 2.2063 0.0269
## DNAmAgeSkinBloodClockAdjAge 0.9364 0.4545 0.0456 1.8272 0.0394
## AgeAccelGrim 0.2992 1.1861 -2.0255 2.6239 0.8008
## DNAmTLAdjAge 0.0160 0.0677 -0.1167 0.1486 0.8135
## IEAA -0.2887 0.6005 -1.4658 0.8883 0.6306
## EEAA 0.4159 0.5071 -0.5780 1.4099 0.4121
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno <= 0.05
## DNAmAgeSkinBloodClockAdjAge <= 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $URI5_CHR
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.0355 0.4677 -0.9522 0.8812 0.9395
## AgeAccelerationResidualHannum 0.0229 0.4457 -0.8506 0.8964 0.9590
## AgeAccelPheno -0.0636 0.3610 -0.7710 0.6439 0.8602
## DNAmAgeSkinBloodClockAdjAge 0.1281 0.3592 -0.5759 0.8321 0.7214
## AgeAccelGrim 0.2868 0.3393 -0.3782 0.9518 0.3979
## DNAmTLAdjAge -0.0001 0.0318 -0.0624 0.0622 0.9964
## IEAA -0.3663 0.4175 -1.1845 0.4519 0.3802
## EEAA 0.1912 0.4998 -0.7885 1.1709 0.7021
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
Sensitivity analyses
Likelihood ratio (LR) test (no confounders)
Full model: \[Y = \beta_0 + \beta_1 *
\text{NAP_1M_2M} + \beta_2 * \text{ACE} + \beta_3 * \text{FLU_PHE} +
\beta_4 * \text{PYR} + \beta_5 * \text{CHR} + \epsilon\]
Nested model: \[Y = \beta_0 +
\epsilon\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.7594 0.9109
## Hannum EAA 0.8236 0.9109
## PhenoAge EAA 0.1275 0.5100
## Skin&Blood EAA 0.5303 0.9109
## GrimAge EAA 0.2542 0.6779
## DNAmTL 0.1227 0.5100
## IEAA 0.8256 0.9109
## EEAA 0.9109 0.9109
Likelihood ratio (LR) test (no confounders) with subjects using only
smoky or smokeless coal
Full model: \[Y = \beta_0 + \beta_1 *
\text{NAP_1M_2M} + \beta_2 * \text{ACE} + \beta_3 * \text{FLU_PHE} +
\beta_4 * \text{PYR} + \beta_5 * \text{CHR} + \epsilon\]
Nested model: \[Y = \beta_0 +
\epsilon\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.5757 0.9426
## Hannum EAA 0.7496 0.9426
## PhenoAge EAA 0.1946 0.9426
## Skin&Blood EAA 0.9001 0.9426
## GrimAge EAA 0.8110 0.9426
## DNAmTL 0.5154 0.9426
## IEAA 0.7304 0.9426
## EEAA 0.9426 0.9426
Likelihood ratio (LR) test (no confounders) with subjects only using
smoky coal
Full model: \[Y = \beta_0 + \beta_1 *
\text{NAP_1M_2M} + \beta_2 * \text{ACE} + \beta_3 * \text{FLU_PHE} +
\beta_4 * \text{PYR} + \beta_5 * \text{CHR} + \epsilon\]
Nested model: \[Y = \beta_0 +
\epsilon\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.7470 0.9344
## Hannum EAA 0.8626 0.9344
## PhenoAge EAA 0.1408 0.9344
## Skin&Blood EAA 0.6584 0.9344
## GrimAge EAA 0.8667 0.9344
## DNAmTL 0.6501 0.9344
## IEAA 0.8187 0.9344
## EEAA 0.9344 0.9344
GEE (No confounders)
In this section, we perform the generalized estimating equations
(GEE) to evaluate the associations between each cluster within current
pollutant exposures and each Epigenetic Age Acceleration with the
formula: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * X + \epsilon
\end{aligned}
\] where \(Y\) is one of the
epigenetic age accelerations and X is one of the cluster estimates.
Results:
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## The estimated effects:
## $URI5_NAP_1M_2M
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.1532 0.4639 -0.7560 1.0625 0.7412
## AgeAccelerationResidualHannum -0.4411 0.4313 -1.2864 0.4043 0.3065
## AgeAccelPheno -0.1046 0.4350 -0.9572 0.7480 0.8100
## DNAmAgeSkinBloodClockAdjAge -0.2763 0.3821 -1.0252 0.4725 0.4695
## AgeAccelGrim 0.6354 0.5447 -0.4322 1.7030 0.2434
## DNAmTLAdjAge -0.0216 0.0156 -0.0521 0.0089 0.1650
## IEAA -0.0941 0.5013 -1.0767 0.8885 0.8511
## EEAA -0.1278 0.5175 -1.1422 0.8865 0.8049
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $URI5_ACE
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.2304 0.5036 -1.2174 0.7566 0.6473
## AgeAccelerationResidualHannum 0.4233 0.3107 -0.1857 1.0324 0.1731
## AgeAccelPheno 0.6801 0.4330 -0.1685 1.5287 0.1162
## DNAmAgeSkinBloodClockAdjAge -0.0808 0.3108 -0.6900 0.5284 0.7949
## AgeAccelGrim 0.7049 0.5222 -0.3187 1.7285 0.1771
## DNAmTLAdjAge -0.0117 0.0111 -0.0336 0.0101 0.2928
## IEAA 0.0695 0.4100 -0.7340 0.8731 0.8653
## EEAA 0.4882 0.4146 -0.3244 1.3009 0.2389
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $URI5_FLU_PHE
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.1892 0.4185 -1.0094 0.6309 0.6511
## AgeAccelerationResidualHannum 0.2082 0.4149 -0.6049 1.0213 0.6158
## AgeAccelPheno 0.3825 0.4086 -0.4184 1.1833 0.3492
## DNAmAgeSkinBloodClockAdjAge 0.0187 0.3207 -0.6098 0.6472 0.9535
## AgeAccelGrim 0.6827 0.4408 -0.1812 1.5466 0.1214
## DNAmTLAdjAge -0.0397 0.0138 -0.0669 -0.0126 0.0041
## IEAA 0.0819 0.4419 -0.7842 0.9480 0.8530
## EEAA 0.4182 0.5515 -0.6627 1.4992 0.4482
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge <= 0.01
## IEAA > 0.05
## EEAA > 0.05
##
## $URI5_PYR
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.7635 0.5926 -1.9250 0.3981 0.1977
## AgeAccelerationResidualHannum 0.3412 0.3616 -0.3675 1.0498 0.3454
## AgeAccelPheno 0.8858 0.4224 0.0579 1.7136 0.0360
## DNAmAgeSkinBloodClockAdjAge 0.5683 0.3935 -0.2030 1.3397 0.1487
## AgeAccelGrim 0.4230 0.2848 -0.1352 0.9812 0.1375
## DNAmTLAdjAge -0.0242 0.0176 -0.0587 0.0103 0.1689
## IEAA -0.3345 0.4314 -1.1800 0.5109 0.4380
## EEAA 0.3930 0.4695 -0.5272 1.3132 0.4025
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno <= 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $URI5_CHR
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.0109 0.3471 -0.6913 0.6695 0.9748
## AgeAccelerationResidualHannum 0.0720 0.3938 -0.6998 0.8439 0.8548
## AgeAccelPheno -0.0272 0.3774 -0.7670 0.7126 0.9425
## DNAmAgeSkinBloodClockAdjAge 0.0536 0.2869 -0.5087 0.6159 0.8518
## AgeAccelGrim 0.4199 0.2632 -0.0960 0.9358 0.1106
## DNAmTLAdjAge -0.0256 0.0202 -0.0653 0.0140 0.2046
## IEAA -0.2227 0.3728 -0.9534 0.5079 0.5502
## EEAA 0.2021 0.4913 -0.7610 1.1651 0.6809
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
GEE (No confounders, mutual adjust)
In this section, we perform the generalized estimating equations
(GEE) to evaluate the associations between each cluster within current
pollutant exposures and each Epigenetic Age Acceleration with the
formula: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * NAP\_1M\_2M + \beta_2 ACE + \beta_3
FLU\_PHE + \beta_4 PYR + \beta_5 CHR \\
& + \epsilon
\end{aligned}
\] where \(Y\) is one of the
epigenetic age accelerations.
Results:
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## [1] "Fitting with 104 observations."
## The estimated effects:
## $URI5_NAP_1M_2M
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.5799 0.5186 -0.4366 1.5964 0.2635
## AgeAccelerationResidualHannum -1.3080 0.5109 -2.3094 -0.3066 0.0105
## AgeAccelPheno -0.6566 0.5159 -1.6677 0.3545 0.2031
## DNAmAgeSkinBloodClockAdjAge -0.4663 0.4402 -1.3290 0.3965 0.2895
## AgeAccelGrim 0.0280 0.5074 -0.9665 1.0225 0.9560
## DNAmTLAdjAge 0.0089 0.0249 -0.0398 0.0576 0.7212
## IEAA -0.2678 0.5757 -1.3963 0.8606 0.6418
## EEAA -0.8372 0.6296 -2.0712 0.3969 0.1836
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum <= 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $URI5_ACE
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.1807 0.4395 -1.0421 0.6808 0.6810
## AgeAccelerationResidualHannum 0.5468 0.2993 -0.0398 1.1333 0.0677
## AgeAccelPheno 0.6156 0.4330 -0.2331 1.4643 0.1551
## DNAmAgeSkinBloodClockAdjAge -0.0957 0.2955 -0.6748 0.4834 0.7460
## AgeAccelGrim 0.4820 0.4113 -0.3241 1.2880 0.2412
## DNAmTLAdjAge -0.0185 0.0173 -0.0524 0.0154 0.2838
## IEAA 0.0973 0.4108 -0.7079 0.9025 0.8128
## EEAA 0.5458 0.4209 -0.2792 1.3707 0.1947
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $URI5_FLU_PHE
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.5462 0.7525 -2.0210 0.9287 0.4680
## AgeAccelerationResidualHannum 1.2976 0.5678 0.1848 2.4104 0.0223
## AgeAccelPheno 0.5314 0.6116 -0.6672 1.7301 0.3848
## DNAmAgeSkinBloodClockAdjAge -0.1065 0.3901 -0.8711 0.6580 0.7848
## AgeAccelGrim 0.3521 0.6006 -0.8251 1.5293 0.5577
## DNAmTLAdjAge -0.0596 0.0404 -0.1387 0.0195 0.1399
## IEAA 0.7401 0.6230 -0.4810 1.9612 0.2348
## EEAA 0.9160 0.7319 -0.5184 2.3505 0.2107
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum <= 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $URI5_PYR
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual -0.4994 0.5663 -1.6094 0.6106 0.3779
## AgeAccelerationResidualHannum 0.0560 0.3586 -0.6468 0.7588 0.8758
## AgeAccelPheno 0.8131 0.4791 -0.1258 1.7521 0.0896
## DNAmAgeSkinBloodClockAdjAge 0.8124 0.4260 -0.0227 1.6474 0.0565
## AgeAccelGrim 0.0904 0.3204 -0.5376 0.7183 0.7779
## DNAmTLAdjAge 0.0110 0.0990 -0.1831 0.2051 0.9117
## IEAA -0.5912 0.5604 -1.6897 0.5072 0.2914
## EEAA 0.0925 0.4799 -0.8481 1.0332 0.8471
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
##
## $URI5_CHR
## coefficient std ci_lower ci_upper p_val
## AgeAccelerationResidual 0.1974 0.6600 -1.0962 1.4910 0.7648
## AgeAccelerationResidualHannum -0.0386 0.4389 -0.8989 0.8217 0.9299
## AgeAccelPheno -0.2237 0.3923 -0.9926 0.5451 0.5685
## DNAmAgeSkinBloodClockAdjAge 0.0808 0.3647 -0.6341 0.7957 0.8247
## AgeAccelGrim 0.2639 0.2049 -0.1377 0.6654 0.1977
## DNAmTLAdjAge 0.0081 0.0601 -0.1096 0.1259 0.8923
## IEAA -0.3579 0.3994 -1.1407 0.4249 0.3702
## EEAA 0.0574 0.5048 -0.9321 1.0469 0.9094
## sig_level
## AgeAccelerationResidual > 0.05
## AgeAccelerationResidualHannum > 0.05
## AgeAccelPheno > 0.05
## DNAmAgeSkinBloodClockAdjAge > 0.05
## AgeAccelGrim > 0.05
## DNAmTLAdjAge > 0.05
## IEAA > 0.05
## EEAA > 0.05
4.1. Current exposure to 5MC
Summary the exposure estimates:
| Characteristic |
Overall, N = 112 |
Smokeles, N = 17 |
Smoky, N = 87 |
Wood_and_or_Plant, N = 8 |
| cur_5mc |
7.81 (5.2, 9.7) |
2.59 (2.2, 4.0) |
9.46 (5.6, 10.1) |
7.40 (7.1, 7.4) |
| (Missing) |
3 |
2 |
1 |
0 |
Primary analysis
Likelihood ratio (LR) test (mix model)
Full model: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * \text{cur_5mc} \\
& + \beta_{2} * county + \beta_{3} * BMI + \beta_{4} * ses +
\beta_{5} * edu + \epsilon
\end{aligned}
\] Nested model: \[
\begin{aligned}
Y = & \beta_0 \\
& + \beta_1 * county + \beta_2 * BMI + \beta_3 * ses + \beta_4 *
edu + \epsilon
\end{aligned}
\] \(H_0\): The full model and
the nested model fit the data equally well. Thus, you should use the
nested model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.6474 0.8024
## Hannum EAA 0.7714 0.8024
## PhenoAge EAA 0.1865 0.4973
## Skin&Blood EAA 0.1241 0.4964
## GrimAge EAA 0.0048 0.0384
## DNAmTL 0.4267 0.8024
## IEAA 0.7782 0.8024
## EEAA 0.8024 0.8024
GEE (mix)
In the following section, we performed generalized estimating
equations (GEE) with equation \[
\begin{aligned}
Y = & \beta_0 + \beta_1 *cur\_5mc\\
& + \beta_2 * county + \beta_3 * BMI + \beta_4 * ses + \beta_5 *
edu + \epsilon
\end{aligned}
\] where \(Y\) is one of the
epigenetic age accelerations.
Results:
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] " Result data: "
## coefficient std ci_lower
## AgeAccelerationResidual 0.043405227 0.115493287924362 -0.18296162
## AgeAccelerationResidualHannum -0.027158185 0.0999991181792323 -0.22315646
## AgeAccelPheno 0.118926749 0.100656276642595 -0.07835955
## DNAmAgeSkinBloodClockAdjAge 0.116943084 0.0716231975184864 -0.02343838
## AgeAccelGrim 0.148955554 0.0506687424790981 0.04964482
## DNAmTLAdjAge -0.004557601 0.00327977322802548 -0.01098596
## IEAA 0.020150560 0.123949043708331 -0.22278957
## EEAA -0.042628589 0.132789328425354 -0.30289567
## ci_upper p_val sig_level
## AgeAccelerationResidual 0.269772072 0.707047242657413 > 0.05
## AgeAccelerationResidualHannum 0.168840087 0.785941715190563 > 0.05
## AgeAccelPheno 0.316213051 0.237398797561661 > 0.05
## DNAmAgeSkinBloodClockAdjAge 0.257324551 0.102520682085509 > 0.05
## AgeAccelGrim 0.248266290 0.00328432839999115 <= 0.01
## DNAmTLAdjAge 0.001870755 0.164647803635041 > 0.05
## IEAA 0.263090685 0.870855969945509 > 0.05
## EEAA 0.217638494 0.748192043476084 > 0.05
## EAAs
## AgeAccelerationResidual Horvath EAA
## AgeAccelerationResidualHannum Hannum EAA
## AgeAccelPheno PhenoAge EAA
## DNAmAgeSkinBloodClockAdjAge Skin&Blood EAA
## AgeAccelGrim GrimAge EAA
## DNAmTLAdjAge DNAmTL
## IEAA IEAA
## EEAA EEAA
Sensitivity analysis
GEE (no confounders)
In the following section, we performed generalized estimating
equations (GEE) with equation \[
\begin{aligned}
Y = & \beta_0 + \beta_1 *cur\_5mc + \epsilon
\end{aligned}
\] where \(Y\) is one of the
epigenetic age accelerations.
Results:
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] " Result data: "
## coefficient std ci_lower
## AgeAccelerationResidual 0.044863339 0.113864633110002 -0.17831134
## AgeAccelerationResidualHannum -0.017041115 0.0978064100987568 -0.20874168
## AgeAccelPheno 0.107159703 0.100283750609095 -0.08939645
## DNAmAgeSkinBloodClockAdjAge 0.106144306 0.0713739765678327 -0.03374869
## AgeAccelGrim 0.140523649 0.0550039747044841 0.03271586
## DNAmTLAdjAge -0.004200799 0.00335993883966329 -0.01078628
## IEAA 0.016148695 0.121516907890369 -0.22202444
## EEAA -0.025686042 0.127591566498255 -0.27576551
## ci_upper p_val sig_level
## AgeAccelerationResidual 0.268038020 0.693576669249562 > 0.05
## AgeAccelerationResidualHannum 0.174659449 0.86168226954879 > 0.05
## AgeAccelPheno 0.303715854 0.285265741749724 > 0.05
## DNAmAgeSkinBloodClockAdjAge 0.246037300 0.136973360441059 > 0.05
## AgeAccelGrim 0.248331440 0.0106251631971735 <= 0.05
## DNAmTLAdjAge 0.002384681 0.211204364462331 > 0.05
## IEAA 0.254321835 0.894278337455743 > 0.05
## EEAA 0.224393428 0.840452605508899 > 0.05
## EAAs
## AgeAccelerationResidual Horvath EAA
## AgeAccelerationResidualHannum Hannum EAA
## AgeAccelPheno PhenoAge EAA
## DNAmAgeSkinBloodClockAdjAge Skin&Blood EAA
## AgeAccelGrim GrimAge EAA
## DNAmTLAdjAge DNAmTL
## IEAA IEAA
## EEAA EEAA
Likelihood ratio (LR) test (no confounders)
Full model: \[
Y = \beta_0 + \beta_1 * cur\_5mc + \epsilon
\]
Nested model: \[Y = \beta_0 +
\epsilon\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.6735 0.8167
## Hannum EAA 0.7749 0.8167
## PhenoAge EAA 0.2570 0.6853
## Skin&Blood EAA 0.1755 0.6853
## GrimAge EAA 0.0124 0.0992
## DNAmTL 0.5136 0.8167
## IEAA 0.8144 0.8167
## EEAA 0.8167 0.8167
Likelihood ratio (LR) test (no confounders) with subjects using only
smoky or smokeless coal
Full model: \[
Y = \beta_0 + \beta_1 * cur\_5mc + \epsilon
\]
Nested model: \[Y = \beta_0 +
\epsilon\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.8934 0.9469
## Hannum EAA 0.5347 0.7711
## PhenoAge EAA 0.4965 0.7711
## Skin&Blood EAA 0.2015 0.7711
## GrimAge EAA 0.0033 0.0264
## DNAmTL 0.5783 0.7711
## IEAA 0.9469 0.9469
## EEAA 0.5483 0.7711
Likelihood ratio (LR) test (no confounders) with subjects only using
smoky coal
Full model: \[
Y = \beta_0 + \beta_1 * cur\_5mc + \epsilon
\]
Nested model: \[Y = \beta_0 +
\epsilon\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.4753 0.7605
## Hannum EAA 0.9663 0.9663
## PhenoAge EAA 0.3272 0.6544
## Skin&Blood EAA 0.2526 0.6544
## GrimAge EAA 0.0084 0.0672
## DNAmTL 0.6585 0.8780
## IEAA 0.3253 0.6544
## EEAA 0.7771 0.8881
4.2. Cumulative exposure to 5MC
Summary the exposure estimates:
| Characteristic |
Overall, N = 112 |
Smokeles, N = 17 |
Smoky, N = 87 |
Wood_and_or_Plant, N = 8 |
| cum_5mc |
253.00 (157.7, 371.9) |
92.43 (82.6, 167.9) |
270.50 (179.9, 389.1) |
341.46 (228.8, 471.5) |
| (Missing) |
3 |
2 |
1 |
0 |
Primary analysis
Likelihood ratio (LR) test (mix model)
Full model: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * \text{cum_5mc} \\
& + \beta_{2} * county + \beta_{3} * BMI + \beta_{4} * ses +
\beta_{5} * edu + \epsilon
\end{aligned}
\] Nested model: \[
\begin{aligned}
Y = & \beta_0 \\
& + \beta_1 * county + \beta_2 * BMI + \beta_3 * ses + \beta_4 *
edu + \epsilon
\end{aligned}
\] \(H_0\): The full model and
the nested model fit the data equally well. Thus, you should use the
nested model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.5554 0.6347
## Hannum EAA 0.2851 0.4562
## PhenoAge EAA 0.0997 0.3988
## Skin&Blood EAA 0.2763 0.4562
## GrimAge EAA 0.0007 0.0056
## DNAmTL 0.5288 0.6347
## IEAA 0.9700 0.9700
## EEAA 0.1615 0.4307
GEE (mix)
In the following section, we performed generalized estimating
equations (GEE) with equation \[
\begin{aligned}
Y = & \beta_0 + \beta_1 *cum\_5mc\\
& + \beta_2 * county + \beta_3 * BMI + \beta_4 * ses + \beta_5 *
edu + \epsilon
\end{aligned}
\] where \(Y\) is one of the
epigenetic age accelerations.
Results:
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] " Result data: "
## coefficient std ci_lower
## AgeAccelerationResidual 0.0011297683 0.00321754110538496 -0.0051766123
## AgeAccelerationResidualHannum 0.0024297203 0.00286859328247939 -0.0031927225
## AgeAccelPheno 0.0046561822 0.00315660422742595 -0.0015307621
## DNAmAgeSkinBloodClockAdjAge 0.0022230277 0.00243998088414274 -0.0025593349
## AgeAccelGrim 0.0058800703 0.00161826592950822 0.0027082691
## DNAmTLAdjAge -0.0001174706 0.000128468642577699 -0.0003692691
## IEAA -0.0007651314 0.00302849757225974 -0.0067009866
## EEAA 0.0042085613 0.00367291833078406 -0.0029903586
## ci_upper p_val sig_level
## AgeAccelerationResidual 0.0074361488 0.725492440526316 > 0.05
## AgeAccelerationResidualHannum 0.0080521632 0.396990881252127 > 0.05
## AgeAccelPheno 0.0108431264 0.140196283843566 > 0.05
## DNAmAgeSkinBloodClockAdjAge 0.0070053902 0.362251089952007 > 0.05
## AgeAccelGrim 0.0090518716 0.000279534738611642 <= 0.001
## DNAmTLAdjAge 0.0001343279 0.360511350051069 > 0.05
## IEAA 0.0051707239 0.8005434220148 > 0.05
## EEAA 0.0114074812 0.251863102495362 > 0.05
## EAAs
## AgeAccelerationResidual Horvath EAA
## AgeAccelerationResidualHannum Hannum EAA
## AgeAccelPheno PhenoAge EAA
## DNAmAgeSkinBloodClockAdjAge Skin&Blood EAA
## AgeAccelGrim GrimAge EAA
## DNAmTLAdjAge DNAmTL
## IEAA IEAA
## EEAA EEAA
Sensitivity analysis
GEE (no confounders)
In the following section, we performed generalized estimating
equations (GEE) with equation \[
\begin{aligned}
Y = & \beta_0 + \beta_1 *cum\_5mc + \epsilon
\end{aligned}
\] where \(Y\) is one of the
epigenetic age accelerations.
Results:
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] " Result data: "
## coefficient std ci_lower
## AgeAccelerationResidual -0.0009917356 0.0027245425973913 -0.0063318390
## AgeAccelerationResidualHannum 0.0024968640 0.00247936234502612 -0.0023626862
## AgeAccelPheno 0.0047965293 0.00266935100058998 -0.0004353987
## DNAmAgeSkinBloodClockAdjAge 0.0013096169 0.00207570564676486 -0.0027587661
## AgeAccelGrim 0.0047733361 0.00163901122252633 0.0015608741
## DNAmTLAdjAge -0.0001203660 0.000108228672285064 -0.0003324941
## IEAA -0.0015487572 0.00257420977692438 -0.0065942084
## EEAA 0.0035090514 0.00318009041422209 -0.0027239258
## ci_upper p_val sig_level
## AgeAccelerationResidual 0.00434836794 0.715857447596787 > 0.05
## AgeAccelerationResidualHannum 0.00735641417 0.313906458923568 > 0.05
## AgeAccelPheno 0.01002845722 0.0723531358161104 > 0.05
## DNAmAgeSkinBloodClockAdjAge 0.00537799999 0.528088827964163 > 0.05
## AgeAccelGrim 0.00798579808 0.00358747245035784 <= 0.01
## DNAmTLAdjAge 0.00009176225 0.266075912628183 > 0.05
## IEAA 0.00349669395 0.547411302915121 > 0.05
## EEAA 0.00974202860 0.269834439221523 > 0.05
## EAAs
## AgeAccelerationResidual Horvath EAA
## AgeAccelerationResidualHannum Hannum EAA
## AgeAccelPheno PhenoAge EAA
## DNAmAgeSkinBloodClockAdjAge Skin&Blood EAA
## AgeAccelGrim GrimAge EAA
## DNAmTLAdjAge DNAmTL
## IEAA IEAA
## EEAA EEAA
Likelihood ratio (LR) test (no confounders)
Full model: \[
Y = \beta_0 + \beta_1 * cum\_5mc + \epsilon
\]
Nested model: \[Y = \beta_0 +
\epsilon\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.7605 0.7605
## Hannum EAA 0.3940 0.6856
## PhenoAge EAA 0.1085 0.4340
## Skin&Blood EAA 0.5849 0.6856
## GrimAge EAA 0.0138 0.1104
## DNAmTL 0.5640 0.6856
## IEAA 0.5999 0.6856
## EEAA 0.3133 0.6856
Likelihood ratio (LR) test (no confounders) with subjects using only
smoky or smokeless coal
Full model: \[
Y = \beta_0 + \beta_1 * cum\_5mc + \epsilon
\]
Nested model: \[Y = \beta_0 +
\epsilon\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.7209 0.7209
## Hannum EAA 0.2799 0.4478
## PhenoAge EAA 0.0759 0.2848
## Skin&Blood EAA 0.1068 0.2848
## GrimAge EAA 0.0280 0.2240
## DNAmTL 0.5272 0.6025
## IEAA 0.5110 0.6025
## EEAA 0.2448 0.4478
Likelihood ratio (LR) test (no confounders) with subjects only using
smoky coal
Full model: \[
Y = \beta_0 + \beta_1 * cum\_5mc + \epsilon
\]
Nested model: \[Y = \beta_0 +
\epsilon\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.9853 0.9853
## Hannum EAA 0.0918 0.2285
## PhenoAge EAA 0.0392 0.1684
## Skin&Blood EAA 0.1428 0.2285
## GrimAge EAA 0.0421 0.1684
## DNAmTL 0.6029 0.8039
## IEAA 0.9819 0.9853
## EEAA 0.1371 0.2285
4.3. Childhood exposure to 5MC
Summary the exposure estimates:
| Characteristic |
Overall, N = 112 |
Smokeles, N = 17 |
Smoky, N = 87 |
Wood_and_or_Plant, N = 8 |
| bir_5mc |
4.83 (2.6, 8.2) |
2.10 (1.5, 4.4) |
4.83 (3.0, 8.2) |
8.02 (3.7, 8.8) |
| (Missing) |
3 |
2 |
1 |
0 |
Primary analysis
Likelihood ratio (LR) test (mix model)
Full model: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * \text{childhood_5mc} \\
& + \beta_{2} * county + \beta_{3} * BMI + \beta_{4} * ses +
\beta_{5} * edu + \epsilon
\end{aligned}
\] Nested model: \[
\begin{aligned}
Y = & \beta_0 \\
& + \beta_1 * county + \beta_2 * BMI + \beta_3 * ses + \beta_4 *
edu + \epsilon
\end{aligned}
\] \(H_0\): The full model and
the nested model fit the data equally well. Thus, you should use the
nested model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.8952 0.9815
## Hannum EAA 0.2856 0.4570
## PhenoAge EAA 0.0385 0.1540
## Skin&Blood EAA 0.1248 0.3328
## GrimAge EAA 0.0009 0.0072
## DNAmTL 0.9815 0.9815
## IEAA 0.4038 0.5384
## EEAA 0.1665 0.3330
GEE (mix)
In the following section, we performed generalized estimating
equations (GEE) with equation \[
\begin{aligned}
Y = & \beta_0 + \beta_1 *childhood\_5mc\\
& + \beta_2 * county + \beta_3 * BMI + \beta_4 * ses + \beta_5 *
edu + \epsilon
\end{aligned}
\] where \(Y\) is one of the
epigenetic age accelerations.
Results:
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] " Result data: "
## coefficient std ci_lower
## AgeAccelerationResidual 0.013581238 0.1525495208842 -0.285415823
## AgeAccelerationResidualHannum 0.135320612 0.143152618792748 -0.145258521
## AgeAccelPheno 0.288191817 0.150828508622891 -0.007432059
## DNAmAgeSkinBloodClockAdjAge 0.156895355 0.112196416684544 -0.063009621
## AgeAccelGrim 0.299554330 0.0856477119470605 0.131684815
## DNAmTLAdjAge -0.001670684 0.00638501299772504 -0.014185309
## IEAA -0.134530357 0.14459026100462 -0.417927269
## EEAA 0.222801828 0.182782972150885 -0.135452798
## ci_upper p_val sig_level
## AgeAccelerationResidual 0.31257830 0.929059351419451 > 0.05
## AgeAccelerationResidualHannum 0.41589974 0.344511294913712 > 0.05
## AgeAccelPheno 0.58381569 0.056039917225708 > 0.05
## DNAmAgeSkinBloodClockAdjAge 0.37680033 0.161993287747863 > 0.05
## AgeAccelGrim 0.46742385 0.000469610751128058 <= 0.001
## DNAmTLAdjAge 0.01084394 0.793585821402502 > 0.05
## IEAA 0.14886655 0.352151209489498 > 0.05
## EEAA 0.58105645 0.222866236215159 > 0.05
## EAAs
## AgeAccelerationResidual Horvath EAA
## AgeAccelerationResidualHannum Hannum EAA
## AgeAccelPheno PhenoAge EAA
## DNAmAgeSkinBloodClockAdjAge Skin&Blood EAA
## AgeAccelGrim GrimAge EAA
## DNAmTLAdjAge DNAmTL
## IEAA IEAA
## EEAA EEAA
Sensitivity analysis
GEE (no confounders)
In the following section, we performed generalized estimating
equations (GEE) with equation \[
\begin{aligned}
Y = & \beta_0 + \beta_1 *childhood\_5mc + \epsilon
\end{aligned}
\] where \(Y\) is one of the
epigenetic age accelerations.
Results:
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] "Fitting with 126 observations."
## [1] " Result data: "
## coefficient std ci_lower
## AgeAccelerationResidual 0.044863339 0.113864633110002 -0.17831134
## AgeAccelerationResidualHannum -0.017041115 0.0978064100987568 -0.20874168
## AgeAccelPheno 0.107159703 0.100283750609095 -0.08939645
## DNAmAgeSkinBloodClockAdjAge 0.106144306 0.0713739765678327 -0.03374869
## AgeAccelGrim 0.140523649 0.0550039747044841 0.03271586
## DNAmTLAdjAge -0.004200799 0.00335993883966329 -0.01078628
## IEAA 0.016148695 0.121516907890369 -0.22202444
## EEAA -0.025686042 0.127591566498255 -0.27576551
## ci_upper p_val sig_level
## AgeAccelerationResidual 0.268038020 0.693576669249562 > 0.05
## AgeAccelerationResidualHannum 0.174659449 0.86168226954879 > 0.05
## AgeAccelPheno 0.303715854 0.285265741749724 > 0.05
## DNAmAgeSkinBloodClockAdjAge 0.246037300 0.136973360441059 > 0.05
## AgeAccelGrim 0.248331440 0.0106251631971735 <= 0.05
## DNAmTLAdjAge 0.002384681 0.211204364462331 > 0.05
## IEAA 0.254321835 0.894278337455743 > 0.05
## EEAA 0.224393428 0.840452605508899 > 0.05
## EAAs
## AgeAccelerationResidual Horvath EAA
## AgeAccelerationResidualHannum Hannum EAA
## AgeAccelPheno PhenoAge EAA
## DNAmAgeSkinBloodClockAdjAge Skin&Blood EAA
## AgeAccelGrim GrimAge EAA
## DNAmTLAdjAge DNAmTL
## IEAA IEAA
## EEAA EEAA
Likelihood ratio (LR) test (no confounders)
Full model: \[
Y = \beta_0 + \beta_1 * childhood\_5mc + \epsilon
\]
Nested model: \[Y = \beta_0 +
\epsilon\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.8964 0.8964
## Hannum EAA 0.4153 0.5537
## PhenoAge EAA 0.0531 0.2124
## Skin&Blood EAA 0.1807 0.4819
## GrimAge EAA 0.0036 0.0288
## DNAmTL 0.8386 0.8964
## IEAA 0.3924 0.5537
## EEAA 0.2927 0.5537
Likelihood ratio (LR) test (no confounders) with subjects using only
smoky or smokeless coal
Full model: \[
Y = \beta_0 + \beta_1 * childhood\_5mc + \epsilon
\]
Nested model: \[Y = \beta_0 +
\epsilon\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.5939 0.6787
## Hannum EAA 0.2654 0.3539
## PhenoAge EAA 0.0603 0.1712
## Skin&Blood EAA 0.0642 0.1712
## GrimAge EAA 0.0064 0.0512
## DNAmTL 0.8264 0.8264
## IEAA 0.1863 0.3213
## EEAA 0.2008 0.3213
Likelihood ratio (LR) test (no confounders) with subjects only using
smoky coal
Full model: \[
Y = \beta_0 + \beta_1 * childhood\_5mc + \epsilon
\]
Nested model: \[Y = \beta_0 +
\epsilon\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.8020 0.8020
## Hannum EAA 0.1088 0.2125
## PhenoAge EAA 0.0409 0.1636
## Skin&Blood EAA 0.0956 0.2125
## GrimAge EAA 0.0079 0.0632
## DNAmTL 0.6425 0.7343
## IEAA 0.4901 0.6535
## EEAA 0.1328 0.2125
5. Ambient Exposure
Summary the exposure estimates:
| Characteristic |
Overall, N = 112 |
Smokeles, N = 17 |
Smoky, N = 87 |
Wood_and_or_Plant, N = 8 |
| bap_air |
39.44 (18.9, 74.1) |
10.09 (4.5, 20.7) |
45.22 (21.9, 76.7) |
69.11 (57.0, 131.2) |
| (Missing) |
4 |
0 |
3 |
1 |
| pm25_air |
139.32 (100.1, 227.1) |
120.16 (102.4, 160.7) |
137.48 (98.3, 211.0) |
421.89 (252.7, 480.4) |
| ANY_air |
564.51 (305.8, 977.5) |
477.86 (187.2, 791.4) |
560.77 (306.0, 914.7) |
7,030.90 (3,125.6, 10,967.7) |
| (Missing) |
35 |
7 |
24 |
4 |
| BPE_air |
46.55 (19.5, 73.4) |
12.70 (3.9, 19.7) |
48.29 (22.9, 83.4) |
66.81 (42.4, 114.8) |
| (Missing) |
4 |
0 |
3 |
1 |
| BaA_air |
40.51 (16.7, 88.1) |
9.44 (2.9, 23.3) |
50.23 (20.7, 106.2) |
68.31 (61.8, 163.2) |
| (Missing) |
4 |
0 |
3 |
1 |
| BbF_air |
62.69 (32.8, 120.9) |
31.76 (13.5, 50.1) |
65.78 (34.5, 124.7) |
88.69 (78.2, 181.6) |
| (Missing) |
4 |
0 |
3 |
1 |
| BkF_air |
13.24 (6.4, 25.9) |
3.37 (2.0, 7.6) |
15.07 (8.0, 28.6) |
27.64 (12.5, 48.0) |
| (Missing) |
4 |
0 |
3 |
1 |
| CHR_air |
45.82 (16.4, 86.9) |
15.24 (4.9, 31.8) |
50.79 (18.1, 86.9) |
91.89 (61.3, 134.8) |
| (Missing) |
4 |
0 |
3 |
1 |
| DBA_air |
12.49 (4.4, 27.5) |
3.92 (1.4, 11.0) |
14.25 (6.1, 31.8) |
12.67 (7.6, 25.3) |
| (Missing) |
4 |
0 |
3 |
1 |
| FLT_air |
17.33 (5.1, 41.6) |
4.35 (0.6, 7.2) |
19.15 (6.5, 41.8) |
104.71 (48.9, 175.2) |
| (Missing) |
4 |
0 |
3 |
1 |
| FLU_air |
276.10 (165.2, 546.9) |
251.42 (219.0, 298.2) |
276.10 (159.0, 544.6) |
1,426.05 (632.8, 2,241.9) |
| (Missing) |
35 |
7 |
24 |
4 |
| IPY_air |
27.29 (14.0, 47.7) |
12.70 (4.3, 16.6) |
30.70 (15.3, 48.1) |
69.17 (51.1, 118.8) |
| (Missing) |
4 |
0 |
3 |
1 |
| NAP_air |
3,170.67 (1,807.5, 5,568.9) |
3,217.69 (2,288.3, 4,623.5) |
3,142.04 (1,759.1, 5,442.8) |
29,828.64 (11,068.1, 49,775.1) |
| (Missing) |
35 |
7 |
24 |
4 |
| PHE_air |
396.14 (220.9, 820.9) |
363.30 (294.3, 550.4) |
380.03 (206.2, 771.8) |
2,120.65 (907.6, 3,404.2) |
| (Missing) |
35 |
7 |
24 |
4 |
| PYR_air |
21.81 (6.1, 51.3) |
6.42 (0.6, 8.2) |
23.96 (7.7, 51.3) |
108.99 (71.5, 191.4) |
| (Missing) |
4 |
0 |
3 |
1 |
Primary analysis
GEE for each ambient exposure measurement (mix model)
In the following section, we performed linear regression with
equation \[
\begin{aligned}
Y = & \beta_0 + \beta_1 *X\\
& + \beta_2 * county + \beta_3 * BMI + \beta_4 * ses + \beta_5 *
edu + \epsilon
\end{aligned}
\] where \(Y\) is one of the
epigenetic age accelerations, and \(X\)
is one of the ambient exposure measurements.
The estimations of \(\beta_1\) with
given \(Y\) and \(X\) are shown below, which can be
interpreted as “the mean of Y changes given a one-unit increase in X
while holding other variables constant”.
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
Likelihood ratio (LR) test (mix model)
Full model: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * \text{bap} + \beta_2 * \text{pm25} +
\beta_3 * \text{ANY} + \beta_4 * \text{BPE} + \beta_5 * \text{BaA} \\
& + \beta_6 * \text{BbF} + \beta_7 * \text{BkF} + \beta_8 *
\text{CHR} + \beta_9 * \text{DBA} + \beta_{10} * \text{FLT} \\
& + \beta_{11} * \text{FLU} + \beta_{12} * \text{IPY} + \beta_{13}
* \text{NAP} + \beta_{14} * \text{PHE} + \beta_{15} * \text{PYR} \\
& + \beta_{16} * county + \beta_{17} * BMI + \beta_{18} * ses +
\beta_{19} * edu + \epsilon
\end{aligned}
\] Nested model: \[
\begin{aligned}
Y = & \beta_0 \\
& + \beta_1 * county + \beta_2 * BMI + \beta_3 * ses + \beta_4 *
edu + \epsilon
\end{aligned}
\] \(H_0\): The full model and
the nested model fit the data equally well. Thus, you should use the
nested model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.0180 0.1440
## Hannum EAA 0.2520 0.2880
## PhenoAge EAA 0.0573 0.1446
## Skin&Blood EAA 0.0797 0.1446
## GrimAge EAA 0.0904 0.1446
## DNAmTL 0.5001 0.5001
## IEAA 0.0538 0.1446
## EEAA 0.2395 0.2880
Sensitivity analysis
GEE for each ambient exposure measurement (no confounders)
In the following section, we performed linear regression with
equation \[
\begin{aligned}
Y = & \beta_0 + \beta_1 *X + \epsilon
\end{aligned}
\] where \(Y\) is one of the
epigenetic age accelerations, and \(X\)
is one of the ambient exposure measurements.
The estimations of \(\beta_1\) with
given \(Y\) and \(X\) are shown below, which can be
interpreted as “the mean of Y changes given a one-unit increase in X
while holding other variables constant”.
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 88 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
## [1] "Fitting with 124 observations."
Likelihood ratio (LR) test (no confounders)
Full model: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * \text{bap} + \beta_2 * \text{pm25} +
\beta_3 * \text{ANY} + \beta_4 * \text{BPE} + \beta_5 * \text{BaA} \\
& + \beta_6 * \text{BbF} + \beta_7 * \text{BkF} + \beta_8 *
\text{CHR} + \beta_9 * \text{DBA} + \beta_{10} * \text{FLT} \\
& + \beta_{11} * \text{FLU} + \beta_{12} * \text{IPY} + \beta_{13}
* \text{NAP} + \beta_{14} * \text{PHE} + \beta_{15} * \text{PYR} \\
& + \epsilon
\end{aligned}
\]
Nested model: \[Y = \beta_0 +
\epsilon\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.0320 0.1160
## Hannum EAA 0.2953 0.3762
## PhenoAge EAA 0.0787 0.1574
## Skin&Blood EAA 0.2369 0.3762
## GrimAge EAA 0.0426 0.1160
## DNAmTL 0.4987 0.4987
## IEAA 0.0435 0.1160
## EEAA 0.3292 0.3762
Likelihood ratio (LR) test (no confounders) with subjects using only
smoky or smokeless coal
Full model: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * \text{bap} + \beta_2 * \text{pm25} +
\beta_3 * \text{ANY} + \beta_4 * \text{BPE} + \beta_5 * \text{BaA} \\
& + \beta_6 * \text{BbF} + \beta_7 * \text{BkF} + \beta_8 *
\text{CHR} + \beta_9 * \text{DBA} + \beta_{10} * \text{FLT} \\
& + \beta_{11} * \text{FLU} + \beta_{12} * \text{IPY} + \beta_{13}
* \text{NAP} + \beta_{14} * \text{PHE} + \beta_{15} * \text{PYR} \\
& + \epsilon
\end{aligned}
\]
Nested model: \[Y = \beta_0 +
\epsilon\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.4697 0.9071
## Hannum EAA 0.7204 0.9071
## PhenoAge EAA 0.5172 0.9071
## Skin&Blood EAA 0.9071 0.9071
## GrimAge EAA 0.1940 0.9071
## DNAmTL 0.8321 0.9071
## IEAA 0.2539 0.9071
## EEAA 0.7178 0.9071
Likelihood ratio (LR) test (no confounders) with subjects only using
smoky coal
Full model: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * \text{bap} + \beta_2 * \text{pm25} +
\beta_3 * \text{ANY} + \beta_4 * \text{BPE} + \beta_5 * \text{BaA} \\
& + \beta_6 * \text{BbF} + \beta_7 * \text{BkF} + \beta_8 *
\text{CHR} + \beta_9 * \text{DBA} + \beta_{10} * \text{FLT} \\
& + \beta_{11} * \text{FLU} + \beta_{12} * \text{IPY} + \beta_{13}
* \text{NAP} + \beta_{14} * \text{PHE} + \beta_{15} * \text{PYR} \\
& + \epsilon
\end{aligned}
\]
Nested model: \[Y = \beta_0 +
\epsilon\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.8076 0.9636
## Hannum EAA 0.8533 0.9636
## PhenoAge EAA 0.3143 0.9636
## Skin&Blood EAA 0.8366 0.9636
## GrimAge EAA 0.0572 0.4576
## DNAmTL 0.9636 0.9636
## IEAA 0.4190 0.9636
## EEAA 0.9106 0.9636
6. Urinary Measurements
Summary the exposure estimates:
| Characteristic |
Overall, N = 112 |
Smokeles, N = 17 |
Smoky, N = 87 |
Wood_and_or_Plant, N = 8 |
| Benzanthracene_Chrysene_urine |
0.38 (0.3, 0.8) |
0.29 (0.1, 0.6) |
0.45 (0.3, 1.0) |
0.36 (0.3, 0.6) |
| (Missing) |
2 |
0 |
2 |
0 |
| Naphthalene_urine |
107.58 (72.1, 168.8) |
96.94 (54.9, 110.9) |
108.85 (73.5, 169.3) |
141.97 (99.7, 174.6) |
| Methylnaphthalene_2_urine |
26.67 (17.9, 45.0) |
17.92 (8.8, 23.4) |
30.18 (20.9, 46.4) |
20.30 (12.2, 34.2) |
| (Missing) |
7 |
0 |
7 |
0 |
| Methylnaphthalene_1_urine |
10.93 (6.6, 18.1) |
5.26 (3.6, 10.5) |
11.52 (7.7, 20.9) |
15.06 (11.0, 26.7) |
| (Missing) |
4 |
1 |
3 |
0 |
| Acenaphthene_urine |
3.14 (2.2, 7.3) |
2.82 (2.2, 3.5) |
3.38 (2.3, 7.9) |
3.58 (2.0, 7.2) |
| Phenanthrene_Anthracene_urine |
112.78 (42.4, 239.6) |
78.75 (41.6, 135.5) |
115.58 (56.8, 239.7) |
109.86 (39.6, 305.8) |
| Fluoranthene_urine |
16.53 (6.1, 23.1) |
17.68 (5.4, 20.8) |
15.25 (6.3, 23.2) |
23.23 (22.4, 36.0) |
| Pyrene_urine |
0.54 (0.4, 0.8) |
0.41 (0.4, 0.4) |
0.54 (0.4, 0.8) |
0.78 (0.7, 0.9) |
| (Missing) |
15 |
7 |
7 |
1 |
Primary analysis
GEE for each ambient exposure measurement (mix)
In the following section, we performed linear regression with
equation \[
\begin{aligned}
Y = & \beta_0 + \beta_1 *X \\
& + \beta_{2} * county + \beta_{3} * BMI + \beta_{4} * ses +
\beta_{5} * edu + \epsilon
\end{aligned}
\] where \(Y\) is one of the
epigenetic age accelerations, and \(X\)
is one of the urinary exposure measurements.
Results:
## [1] "Fitting with 127 observations."
## [1] "Fitting with 127 observations."
## [1] "Fitting with 127 observations."
## [1] "Fitting with 127 observations."
## [1] "Fitting with 127 observations."
## [1] "Fitting with 127 observations."
## [1] "Fitting with 127 observations."
## [1] "Fitting with 127 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 120 observations."
## [1] "Fitting with 120 observations."
## [1] "Fitting with 120 observations."
## [1] "Fitting with 120 observations."
## [1] "Fitting with 120 observations."
## [1] "Fitting with 120 observations."
## [1] "Fitting with 120 observations."
## [1] "Fitting with 120 observations."
## [1] "Fitting with 125 observations."
## [1] "Fitting with 125 observations."
## [1] "Fitting with 125 observations."
## [1] "Fitting with 125 observations."
## [1] "Fitting with 125 observations."
## [1] "Fitting with 125 observations."
## [1] "Fitting with 125 observations."
## [1] "Fitting with 125 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 111 observations."
## [1] "Fitting with 111 observations."
## [1] "Fitting with 111 observations."
## [1] "Fitting with 111 observations."
## [1] "Fitting with 111 observations."
## [1] "Fitting with 111 observations."
## [1] "Fitting with 111 observations."
## [1] "Fitting with 111 observations."
Likelihood ratio (LR) test (mix model)
Full model: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * \text{Benzanthracene_Chrysene} + \beta_2
* \text{Naphthalene} \\
& + \beta_3 * \text{2.Methylnaphthalene} + \beta_4 *
\text{1.Methylnaphthalene} \\
& + \beta_5 * \text{Acenaphthene }+ \beta_6 *
\text{Phenanthrene_Anthracene} \\
& + \beta_7 * \text{Phenanthrene_Anthracene} + \beta_8 *
\text{Fluoranthene} + \beta_9 * \text{Pyrene} \\
& + \beta_{10} * county + \beta_{11} * BMI + \beta_{12} * ses +
\beta_{13} * edu + \epsilon
\end{aligned}
\] Nested model: \[
\begin{aligned}
Y = & \beta_0 \\
& + \beta_1 * county + \beta_2 * BMI + \beta_3 * ses + \beta_4 *
edu + \epsilon
\end{aligned}
\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.0014 0.0056
## Hannum EAA 0.0643 0.0735
## PhenoAge EAA 0.0007 0.0056
## Skin&Blood EAA 0.0048 0.0077
## GrimAge EAA 0.0041 0.0077
## DNAmTL 0.0172 0.0229
## IEAA 0.0046 0.0077
## EEAA 0.1926 0.1926
Sensitivity analysis
GEE for each ambient exposure measurement (no confounders)
In the following section, we performed linear regression with
equation \[
\begin{aligned}
Y = & \beta_0 + \beta_1 *X + \epsilon
\end{aligned}
\] where \(Y\) is one of the
epigenetic age accelerations, and \(X\)
is one of the urinary exposure measurements.
Results:
## [1] "Fitting with 127 observations."
## [1] "Fitting with 127 observations."
## [1] "Fitting with 127 observations."
## [1] "Fitting with 127 observations."
## [1] "Fitting with 127 observations."
## [1] "Fitting with 127 observations."
## [1] "Fitting with 127 observations."
## [1] "Fitting with 127 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 120 observations."
## [1] "Fitting with 120 observations."
## [1] "Fitting with 120 observations."
## [1] "Fitting with 120 observations."
## [1] "Fitting with 120 observations."
## [1] "Fitting with 120 observations."
## [1] "Fitting with 120 observations."
## [1] "Fitting with 120 observations."
## [1] "Fitting with 125 observations."
## [1] "Fitting with 125 observations."
## [1] "Fitting with 125 observations."
## [1] "Fitting with 125 observations."
## [1] "Fitting with 125 observations."
## [1] "Fitting with 125 observations."
## [1] "Fitting with 125 observations."
## [1] "Fitting with 125 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 129 observations."
## [1] "Fitting with 111 observations."
## [1] "Fitting with 111 observations."
## [1] "Fitting with 111 observations."
## [1] "Fitting with 111 observations."
## [1] "Fitting with 111 observations."
## [1] "Fitting with 111 observations."
## [1] "Fitting with 111 observations."
## [1] "Fitting with 111 observations."
Likelihood ratio (LR) test (no confounders)
Full model: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * \text{Benzanthracene_Chrysene} + \beta_2
* \text{Naphthalene} \\
& + \beta_3 * \text{2.Methylnaphthalene} + \beta_4 *
\text{1.Methylnaphthalene} \\
& + \beta_5 * \text{Acenaphthene }+ \beta_6 *
\text{Phenanthrene_Anthracene} \\
& + \beta_7 * \text{Phenanthrene_Anthracene} + \beta_8 *
\text{Fluoranthene} + \beta_9 * \text{Pyrene} \\
& + \epsilon
\end{aligned}
\] Nested model: \[Y = \beta_0 +
\epsilon\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.0031 0.0124
## Hannum EAA 0.0807 0.0922
## PhenoAge EAA 0.0029 0.0124
## Skin&Blood EAA 0.0063 0.0168
## GrimAge EAA 0.0264 0.0352
## DNAmTL 0.0243 0.0352
## IEAA 0.0111 0.0222
## EEAA 0.1922 0.1922
Likelihood ratio (LR) test (no confounders) with subjects using only
smoky or smokeless coal
Full model: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * \text{Benzanthracene_Chrysene} + \beta_2
* \text{Naphthalene} \\
& + \beta_3 * \text{2.Methylnaphthalene} + \beta_4 *
\text{1.Methylnaphthalene} \\
& + \beta_5 * \text{Acenaphthene }+ \beta_6 *
\text{Phenanthrene_Anthracene} \\
& + \beta_7 * \text{Phenanthrene_Anthracene} + \beta_8 *
\text{Fluoranthene} + \beta_9 * \text{Pyrene} \\
& + \epsilon
\end{aligned}
\] Nested model: \[Y = \beta_0 +
\epsilon\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.0278 0.0556
## Hannum EAA 0.1242 0.1945
## PhenoAge EAA 0.0068 0.0272
## Skin&Blood EAA 0.0008 0.0064
## GrimAge EAA 0.3939 0.3939
## DNAmTL 0.1600 0.1945
## IEAA 0.0153 0.0408
## EEAA 0.1702 0.1945
Likelihood ratio (LR) test (no confounders) with subjects only using
smoky coal
Full model: \[
\begin{aligned}
Y = & \beta_0 + \beta_1 * \text{Benzanthracene_Chrysene} + \beta_2
* \text{Naphthalene} \\
& + \beta_3 * \text{2.Methylnaphthalene} + \beta_4 *
\text{1.Methylnaphthalene} \\
& + \beta_5 * \text{Acenaphthene }+ \beta_6 *
\text{Phenanthrene_Anthracene} \\
& + \beta_7 * \text{Phenanthrene_Anthracene} + \beta_8 *
\text{Fluoranthene} + \beta_9 * \text{Pyrene} \\
& + \epsilon
\end{aligned}
\] Nested model: \[Y = \beta_0 +
\epsilon\]
\(H_0\): The full model and the
nested model fit the data equally well. Thus, you should use the nested
model.
\(H_A\): The full model fits the data
significantly better than the nested model. Thus, you should use the
full model.
P-values results:
## p_vals p_vals_BHadj
## Horvath EAA 0.3183 0.4334
## Hannum EAA 0.3792 0.4334
## PhenoAge EAA 0.0270 0.1080
## Skin&Blood EAA 0.1362 0.3632
## GrimAge EAA 0.4765 0.4765
## DNAmTL 0.2149 0.4298
## IEAA 0.0242 0.1080
## EEAA 0.3477 0.4334